WorkingFolder/PythonCode/PrepareParameters.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pickle\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numba as nb\n",
    "from copy import copy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\TaoWang\\AppData\\Local\\Temp\\ipykernel_15164\\2957389194.py:4: MatplotlibDeprecationWarning: The seaborn styles shipped by Matplotlib are deprecated since 3.6, as they no longer correspond to the styles shipped by seaborn. However, they will remain available as 'seaborn-v0_8-<style>'. Alternatively, directly use the seaborn API instead.\n",
      "  plt.style.use('seaborn')\n"
     ]
    }
   ],
   "source": [
    "## figure plotting configurations\n",
    "\n",
    "\n",
    "plt.style.use('seaborn')\n",
    "plt.rcParams[\"font.family\"] = \"Times New Roman\" #'serif'\n",
    "plt.rcParams['font.serif'] = 'Ubuntu'\n",
    "plt.rcParams['font.monospace'] = 'Ubuntu Mono'\n",
    "plt.rcParams['axes.labelweight'] = 'bold'\n",
    "\n",
    "## Set the plotting style\n",
    "plt.rc('font', size=25)\n",
    "# Set the axes title font size\n",
    "plt.rc('axes', titlesize=20)\n",
    "# Set the axes labels font size\n",
    "plt.rc('axes', labelsize=20)\n",
    "# Set the font size for x tick labels\n",
    "plt.rc('xtick', labelsize=20)\n",
    "# Set the font size for y tick labels\n",
    "plt.rc('ytick', labelsize=20)\n",
    "# Set the legend font size\n",
    "plt.rc('legend', fontsize=20)\n",
    "# Set the font size of the figure title\n",
    "plt.rc('figure', titlesize=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "## some life cycle paras \n",
    "T = 40\n",
    "L = 60\n",
    "\n",
    "T_q = T*4\n",
    "L_q = L*4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "def y2q_interpolate(xs_y):\n",
    "    \n",
    "    \"\"\"\n",
    "    this function turns an array of yealry rates into quarterly rates via linear interpolation \n",
    "    \"\"\"\n",
    "\n",
    "    n_y = len(xs_y)\n",
    "    y_id = np.arange(n_y)\n",
    "    q_id = y_id*4\n",
    "\n",
    "    n_q = 4*n_y\n",
    "    xs_q = np.empty(n_q)\n",
    "    \n",
    "    for i_y in range(n_y):\n",
    "        xs_q[4*i_y] = xs_y[i_y]\n",
    "        for i_q in np.arange(1,4):\n",
    "            q_id_this = 4*i_y+i_q\n",
    "            xs_q[q_id_this] = np.interp(q_id_this,q_id,xs_y)\n",
    "            \n",
    "    return xs_q"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Survival probability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "constant_survial = True\n",
    "\n",
    "#https://www.ssa.gov/oact/STATS/table4c6.html\n",
    "life_table = pd.read_excel(\"../OtherData/lifetable.xlsx\",sheet_name='data')\n",
    "life_table['av_survial'] = 1-(life_table['male_D']+life_table['female_D'])/2\n",
    "LivProbs_y = np.array(life_table[(life_table['age']>25)&(life_table['age']<=25+L)]['av_survial'])\n",
    "LivProbs_q = y2q_interpolate(LivProbs_y)\n",
    "                                            \n",
    "if constant_survial:\n",
    "    LivProbs_y = np.ones_like(LivProbs_y)*(1-0.006)\n",
    "    LivProbs_q = y2q_interpolate(LivProbs_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Population growth rate \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "pop_n = 0.005 ## population growth rate from U.S. census"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Age profile of income "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "code_folding": [
     6,
     39
    ],
    "lines_to_next_cell": 2
   },
   "outputs": [],
   "source": [
    "#############################################\n",
    "### Choose the data source of age profile here\n",
    "#############################################\n",
    "\n",
    "age_profile_data ='SIPP'\n",
    "\n",
    "if age_profile_data=='SIPP':\n",
    "    ## import age income profile \n",
    "\n",
    "    age_profile = pd.read_stata('../OtherData/age_profile.dta')   \n",
    "\n",
    "    ## select age range for the model and turn it into an array \n",
    "    lc_wages = np.array(age_profile[(age_profile['age']>=24) &(age_profile['age']<=24+T)]['wage_age'])\n",
    "    #print(str(len(lc_wages)),'years since age 25')\n",
    "\n",
    "    ## growth rates since initial age over life cicle before retirement\n",
    "    lc_G = lc_wages[1:]/lc_wages[:-1]\n",
    "\n",
    "    ## growth rates after retirement\n",
    "\n",
    "    lc_G_rt = np.ones(L-len(lc_G)) ## fill the rest periods \n",
    "    lc_G_rt[0] = 1/np.cumprod(lc_G)[-1]\n",
    "\n",
    "\n",
    "    lc_G_full = np.concatenate([lc_G,lc_G_rt])\n",
    "    assert len(lc_G_full) == L,'length of G needs to be equal to L'\n",
    "    #lc_G_full = np.ones_like(lc_G_full)\n",
    "\n",
    "    \n",
    "    ## turn yearly number to quarterly number with interpolation \n",
    "\n",
    "    ## get the quarterly income profile over life cycle before retirement \n",
    "    lc_G_q = y2q_interpolate(lc_G)\n",
    "\n",
    "    lc_G_q_rt = y2q_interpolate(lc_G_rt)\n",
    "\n",
    "    lc_G_q_full = np.concatenate([lc_G_q,lc_G_q_rt])\n",
    "    \n",
    "    \n",
    "elif age_profile_data=='SCF':\n",
    "    ## import age income profile \n",
    "    SCF_profile = pd.read_pickle('data/SCF_age_profile.pkl')\n",
    "    SCF_profile = SCF_profile[(SCF_profile.index>=24) & (SCF_profile.index<=24+T)]\n",
    "    ## make age profile using age polynomial regressions \n",
    "    \n",
    "    import statsmodels.api as sm\n",
    "    # Extract age and lwage_income from the data Series\n",
    "    age = SCF_profile.index+25\n",
    "    lwage_income = SCF_profile['av_lwage_income']\n",
    "        \n",
    "    # Create a design matrix with polynomial features\n",
    "    X = np.column_stack((age, age**2,age**3,age**4))  # Third-order polynomial\n",
    "    # Add a constant term for the intercept\n",
    "    X = sm.add_constant(X)\n",
    "    # Fit the OLG-regression model\n",
    "    ols_results = sm.OLS(lwage_income, X).fit()\n",
    "    # Predict values using the fitted model\n",
    "    predicted_values = ols_results.predict(X)\n",
    "    SCF_profile['av_lwage_income_pr'] = predicted_values\n",
    "\n",
    "    lc_p_incom = np.exp(np.array(SCF_profile['av_lwage_income_pr']))\n",
    "    lc_G = lc_p_incom[1:]/lc_p_incom[:-1]\n",
    "    \n",
    "    lc_G_rt = np.ones(L-len(lc_G))\n",
    "    lc_G_rt[0] = 1/np.cumprod(lc_G)[-1]\n",
    "\n",
    "    lc_G_full = np.concatenate([lc_G,lc_G_rt])\n",
    "    assert len(lc_G_full) == L,'length of G needs to be equal to L'\n",
    "    \n",
    "    lc_G_q_full = y2q_interpolate(lc_G_full)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "code_folding": [
     0,
     2
    ]
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if __name__ == \"__main__\":\n",
    "    plt.title('Determinstic life-cycle wage profile')\n",
    "    plt.plot(np.cumprod(lc_G_full),\n",
    "            'k-v')\n",
    "    plt.xlabel('Working Age')\n",
    "    plt.ylabel('Wage Relative to 25yr old')\n",
    "    plt.tight_layout()\n",
    "    plt.savefig('../Graphs/sipp/age_wage_profile.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "## subjective growth expectations \n",
    "\n",
    "SCE = pd.read_stata('../SurveyData/SCE/IncExpSCEProbIndM.dta')   \n",
    "SCE = SCE.rename(columns={'Q24_mean': 'incexp',\n",
    "                       'Q24_var': 'incvar',\n",
    "                       'Q24_iqr': 'inciqr',\n",
    "                       'Q24_rmean':'rincexp',\n",
    "                       'Q24_rvar': 'rincvar',\n",
    "                       'Q13new':'UE_s',\n",
    "                       'Q22new':'UE_f'\n",
    "                       })\n",
    "SCE = SCE.rename(columns = {'D6':'HHinc',\n",
    "                          'Q10_1':'fulltime',\n",
    "                          'Q10_2':'parttime',\n",
    "                          'Q12new':'selfemp',\n",
    "                          'Q32':'age',\n",
    "                          'Q33':'gender',\n",
    "                          'Q36':'educ'})\n",
    "\n",
    "lc_G_sub = np.array(1+SCE[(SCE['age']>=26) &(SCE['age']<64)].groupby('age')['rincexp'].mean())\n",
    "## growth rates after retirement\n",
    "lc_G_rt_sub = np.ones(L-T)\n",
    "#lc_G_rt[0] = 1/np.cumprod(lc_G)[-1]\n",
    "#lc_G_rt = lc_G_rt*lc_G[-1]\n",
    "lc_G_full_sub = np.concatenate([lc_G_sub,lc_G_rt_sub])\n",
    "assert len(lc_G_full_sub) == L,'length of G needs to be equal to L'\n",
    "lc_G_q_full_sub = y2q_interpolate(lc_G_full_sub)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if __name__ == \"__main__\":\n",
    "    plt.title('subjective life-cycle profile')\n",
    "    plt.plot(np.cumprod(lc_G_full_sub))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Income risk estimates "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "risks_est = pd.read_stata('../OtherData/sipp/sipp_history_vol_decomposed.dta')\n",
    "## risks of permanent and transitory component \n",
    "\n",
    "σ_ψ_q_sipp = np.sqrt(risks_est['permanent']**2*3)\n",
    "σ_θ_q_sipp = np.sqrt(risks_est['permanent']**2/3)\n",
    "\n",
    "## p/t ratio \n",
    "kappas_sipp  = risks_est['permanent']/risks_est['transitory']\n",
    "kappa_sipp = np.median(kappas_sipp.dropna())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "## risks of permanent and transitory component \n",
    "\n",
    "σ_ψ_q_sipp = np.sqrt(risks_est['permanent']**2*3)\n",
    "σ_θ_q_sipp = np.sqrt(risks_est['permanent']**2/3)\n",
    "\n",
    "## p/t ratio \n",
    "kappas_sipp  = risks_est['permanent']/risks_est['transitory']\n",
    "kappa_sipp = np.median(kappas_sipp.dropna())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initial conditions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.stats.weightstats import DescrStatsW\n",
    "\n",
    "## SCF data \n",
    "SCF2016 = pd.read_stata('rscfp2016.dta')\n",
    "SCF2016 = SCF2016[SCF2016['age']==25]\n",
    "SCF2016 = SCF2016.drop_duplicates(subset=['yy1'])\n",
    "SCF2016 = SCF2016[['norminc','networth','wgt']]\n",
    "SCF2016 = SCF2016[SCF2016['norminc']>0]\n",
    "\n",
    "## permanent income at age 25 \n",
    "pinc_SCF = SCF2016[['norminc','wgt']]\n",
    "lpinc_SCF_pos, lpinc_SCF_wgt = np.log(pinc_SCF['norminc']),np.array(pinc_SCF['wgt'])\n",
    "\n",
    "## wealth \n",
    "b_SCF = SCF2016[['networth','wgt']]\n",
    "lb_SCF_pos, lb_SCF_wgt = b_SCF['networth'],np.array(b_SCF['wgt'])\n",
    "\n",
    "## wealth to permanent income\n",
    "SCF2016['networth2pinc'] = SCF2016['networth']/SCF2016['norminc']\n",
    "b_SCF, b_wgt = SCF2016['networth2pinc'], SCF2016['wgt']\n",
    "\n",
    "## compute data moments \n",
    "σ_ψ_init_SCF = DescrStatsW(lpinc_SCF_pos, weights=lpinc_SCF_wgt, ddof=1).std\n",
    "σ_b_init_SCF = DescrStatsW(lb_SCF_pos, weights=lb_SCF_wgt, ddof=1).std\n",
    "b_SCF = DescrStatsW(b_SCF,weights=b_wgt, ddof=1).mean ## annual \n",
    "b_q_SCF = b_SCF*4  ## quarterly  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### subjective profile estiamtion "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "perceived transitory PR: 0.02055997279522646\n",
      "perceived permanent PR: 0.034311623666908396\n"
     ]
    }
   ],
   "source": [
    "## subjective constant profile \n",
    "\n",
    "sigma_eps_sub = np.sqrt(1/(1+kappa_sipp**2)*0.04**2) ## 0.04 is an upper bound for the average PR in standard deviation from SCE \n",
    "sigma_psi_sub = sigma_eps_sub*kappa_sipp\n",
    "\n",
    "if __name__ == \"__main__\":    \n",
    "    print('perceived transitory PR:',sigma_eps_sub)\n",
    "    print('perceived permanent PR:',sigma_psi_sub)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "## subjective markov profile \n",
    "\n",
    "\n",
    "## import subjective profile estimation results \n",
    "SCE_est_q = pd.read_pickle('data/subjective_profile_est_q.pkl')\n",
    "SCE_est_y = pd.read_pickle('data/subjective_profile_est_y.pkl')\n",
    "SCE_est_q = SCE_est_q['baseline']\n",
    "SCE_est_y = SCE_est_y['baseline']\n",
    "P_sub_q = np.array([[SCE_est_q.loc['$q$'],1-SCE_est_q.loc['$q$']],\n",
    "                    [1-SCE_est_q.loc['$p$'],SCE_est_q.loc['$p$']]])\n",
    "P_sub_y = np.array([[SCE_est_y.loc['$q$'],1-SCE_est_y.loc['$q$']],\n",
    "                    [1-SCE_est_y.loc['$p$'],SCE_est_y.loc['$p$']]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "$std(\\tilde\\sigma)$        1.189659\n",
       "$q$                        0.975694\n",
       "$p$                        0.975694\n",
       "$\\tilde\\sigma^l_\\psi$      0.487531\n",
       "$\\tilde\\sigma^l_\\theta$    0.092222\n",
       "$\\tilde\\sigma^h_\\psi$      1.391305\n",
       "$\\tilde\\sigma^h_\\theta$    0.263179\n",
       "Name: baseline, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SCE_est_q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "## create a dictionary of parameters \n",
    "life_cycle_paras_q = {'ρ': 2.0, \n",
    "                    'β': 0.97**(1/4), \n",
    "                    'P': np.array([[0.18, 0.82],\n",
    "                                   [0.04, 0.96]]), \n",
    "                    'z_val': np.array([0., 1.]), \n",
    "                    'σ_ψ': np.sqrt(0.15**2*4/11), \n",
    "                    'σ_θ': np.sqrt(0.15**2*4), \n",
    "                    'U': 0.0, \n",
    "                    'LivPrb': LivProbs_q, \n",
    "                    'n': (1+pop_n)**(1/4)-1,\n",
    "                    'R': 1.01**(1/4), \n",
    "                    'W': 1.0, \n",
    "                    'T': T_q, \n",
    "                    'L': L_q, \n",
    "                    'G':lc_G_q_full, \n",
    "                    'G_sub':lc_G_q_full_sub, \n",
    "                    'unemp_insurance': 0.15, \n",
    "                    'pension': 0.65, \n",
    "                    'σ_ψ_init': σ_ψ_init_SCF, \n",
    "                    'init_b': b_q_SCF, \n",
    "                    'λ': 0.0, \n",
    "                    'λ_SS': 0.065, \n",
    "                    'transfer': 0.0, \n",
    "                    'bequest_ratio': 0.0,\n",
    "                    'κ':kappa_sipp,\n",
    "                     ## subjective profile\n",
    "                    'σ_ψ_sub':sigma_psi_sub**2*4/11,\n",
    "                    'σ_θ_sub':sigma_eps_sub**2*4,\n",
    "                    'P_sub': P_sub_q,\n",
    "                    'σ_ψ_2mkv':np.array([SCE_est_q.loc['$\\tilde\\sigma^l_\\psi$'],\n",
    "                                       SCE_est_q.loc['$\\tilde\\sigma^h_\\psi$']]),\n",
    "                    'σ_θ_2mkv':np.array([SCE_est_q.loc['$\\tilde\\sigma^l_\\theta$'],\n",
    "                                       SCE_est_q.loc['$\\tilde\\sigma^h_\\theta$']])\n",
    "                }\n",
    "\n",
    "\n",
    "if '$\\tilde \\mho^l$' in SCE_est_q.index:\n",
    "\n",
    "    life_cycle_paras_q['mho_2mkv'] = np.array([SCE_est_q.loc['$\\tilde \\mho^l$'],\n",
    "                                             SCE_est_q.loc['$\\tilde \\mho^h$']])\n",
    "    life_cycle_paras_q['E_2mkv'] = np.array([SCE_est_q.loc['$\\tilde E^l$'],\n",
    "                                          SCE_est_q.loc['$\\tilde E^h$']])\n",
    "    \n",
    "else:\n",
    "    life_cycle_paras_q['mho_2mkv'] = np.array([life_cycle_paras_q['P'][0,0],\n",
    "                                               life_cycle_paras_q['P'][0,0]])\n",
    "    life_cycle_paras_q['E_2mkv'] = np.array([life_cycle_paras_q['P'][1,1],\n",
    "                                           life_cycle_paras_q['P'][1,1]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ρ': 2.0,\n",
       " 'β': 0.9924141172814958,\n",
       " 'P': array([[0.18, 0.82],\n",
       "        [0.04, 0.96]]),\n",
       " 'z_val': array([0., 1.]),\n",
       " 'σ_ψ': 0.09045340337332908,\n",
       " 'σ_θ': 0.3,\n",
       " 'U': 0.0,\n",
       " 'LivPrb': array([0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994, 0.994,\n",
       "        0.994, 0.994, 0.994, 0.994, 0.994, 0.994]),\n",
       " 'n': 0.0012476630625266605,\n",
       " 'R': 1.0024906793143211,\n",
       " 'W': 1.0,\n",
       " 'T': 160,\n",
       " 'L': 240,\n",
       " 'G': array([1.02716959, 1.02675298, 1.02633637, 1.02591977, 1.02550316,\n",
       "        1.0250954 , 1.02468765, 1.02427989, 1.02387214, 1.02347335,\n",
       "        1.02307457, 1.02267578, 1.022277  , 1.02188739, 1.02149779,\n",
       "        1.02110818, 1.02071857, 1.02033839, 1.0199582 , 1.01957801,\n",
       "        1.01919782, 1.01882711, 1.0184564 , 1.01808569, 1.01771498,\n",
       "        1.01735398, 1.01699299, 1.01663199, 1.016271  , 1.01591989,\n",
       "        1.01556879, 1.01521769, 1.01486659, 1.01452556, 1.01418453,\n",
       "        1.01384351, 1.01350248, 1.01317167, 1.01284087, 1.01251006,\n",
       "        1.01217926, 1.01185888, 1.01153851, 1.01121813, 1.01089776,\n",
       "        1.01058793, 1.01027811, 1.00996828, 1.00965846, 1.00935945,\n",
       "        1.00906044, 1.00876144, 1.00846243, 1.00817433, 1.00788623,\n",
       "        1.00759813, 1.00731003, 1.00703308, 1.00675613, 1.00647917,\n",
       "        1.00620222, 1.0059365 , 1.00567079, 1.00540507, 1.00513935,\n",
       "        1.00488517, 1.00463098, 1.0043768 , 1.00412261, 1.00388008,\n",
       "        1.00363755, 1.00339502, 1.00315249, 1.00292176, 1.00269103,\n",
       "        1.0024603 , 1.00222957, 1.00201088, 1.00179219, 1.0015735 ,\n",
       "        1.00135481, 1.00114828, 1.00094175, 1.00073522, 1.00052869,\n",
       "        1.00033461, 1.00014052, 0.99994643, 0.99975234, 0.99957074,\n",
       "        0.99938914, 0.99920754, 0.99902594, 0.99885711, 0.99868828,\n",
       "        0.99851945, 0.99835062, 0.99819474, 0.99803886, 0.99788298,\n",
       "        0.9977271 , 0.9975843 , 0.9974415 , 0.9972987 , 0.9971559 ,\n",
       "        0.99702646, 0.99689701, 0.99676757, 0.99663812, 0.99652214,\n",
       "        0.99640617, 0.99629019, 0.99617422, 0.99607193, 0.99596965,\n",
       "        0.99586737, 0.99576509, 0.99567673, 0.99558836, 0.9955    ,\n",
       "        0.99541163, 0.99533731, 0.99526298, 0.99518865, 0.99511433,\n",
       "        0.99505433, 0.99499434, 0.99493435, 0.99487436, 0.99482881,\n",
       "        0.99478325, 0.9947377 , 0.99469215, 0.99466129, 0.99463043,\n",
       "        0.99459957, 0.99456871, 0.99455273, 0.99453676, 0.99452078,\n",
       "        0.99450481, 0.99450397, 0.99450314, 0.99450231, 0.99450147,\n",
       "        0.9945159 , 0.99453032, 0.99454474, 0.99455917, 0.99458915,\n",
       "        0.99461913, 0.99464911, 0.99467909, 0.99472484, 0.99477059,\n",
       "        0.99481633, 0.99486208, 0.99486208, 0.99486208, 0.99486208,\n",
       "        0.82025215, 0.86518911, 0.91012607, 0.95506304, 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ]),\n",
       " 'G_sub': array([1.00601101, 1.00499138, 1.00397176, 1.00295213, 1.0019325 ,\n",
       "        1.00219724, 1.00246197, 1.0027267 , 1.00299144, 1.00207262,\n",
       "        1.0011538 , 1.00023498, 0.99931616, 0.99947621, 0.99963626,\n",
       "        0.99979632, 0.99995637, 1.00027534, 1.00059432, 1.00091329,\n",
       "        1.00123227, 1.00011246, 0.99899265, 0.99787284, 0.99675304,\n",
       "        0.99709232, 0.99743161, 0.99777089, 0.99811018, 0.99809729,\n",
       "        0.9980844 , 0.99807151, 0.99805862, 0.99724846, 0.99643829,\n",
       "        0.99562813, 0.99481797, 0.99581884, 0.9968197 , 0.99782057,\n",
       "        0.99882144, 0.99797761, 0.99713379, 0.99628997, 0.99544615,\n",
       "        0.9924615 , 0.98947686, 0.98649222, 0.98350757, 0.98679487,\n",
       "        0.99008217, 0.99336948, 0.99665678, 0.99597934, 0.9953019 ,\n",
       "        0.99462447, 0.99394703, 0.99401282, 0.99407861, 0.99414439,\n",
       "        0.99421018, 0.99407102, 0.99393186, 0.9937927 , 0.99365354,\n",
       "        0.9935872 , 0.99352086, 0.99345452, 0.99338818, 0.9935547 ,\n",
       "        0.99372122, 0.99388774, 0.99405426, 0.9939829 , 0.99391153,\n",
       "        0.99384017, 0.99376881, 0.99342866, 0.99308851, 0.99274836,\n",
       "        0.99240822, 0.99195017, 0.99149212, 0.99103408, 0.99057603,\n",
       "        0.99044226, 0.99030849, 0.99017473, 0.99004096, 0.99078849,\n",
       "        0.99153602, 0.99228355, 0.99303108, 0.98526108, 0.97749108,\n",
       "        0.96972108, 0.96195108, 0.96862221, 0.97529334, 0.98196447,\n",
       "        0.9886356 , 0.98881343, 0.98899126, 0.98916909, 0.98934692,\n",
       "        0.98860668, 0.98786643, 0.98712619, 0.98638594, 0.98686045,\n",
       "        0.98733497, 0.98780948, 0.98828399, 0.98790744, 0.98753089,\n",
       "        0.98715433, 0.98677778, 0.98795986, 0.98914194, 0.99032402,\n",
       "        0.9915061 , 0.99045476, 0.98940343, 0.98835209, 0.98730075,\n",
       "        0.98713878, 0.9869768 , 0.98681483, 0.98665285, 0.98712841,\n",
       "        0.98760396, 0.98807952, 0.98855507, 0.98924834, 0.9899416 ,\n",
       "        0.99063486, 0.99132812, 0.98969367, 0.98805922, 0.98642477,\n",
       "        0.98479033, 0.98572953, 0.98666874, 0.98760794, 0.98854715,\n",
       "        0.9885321 , 0.98851705, 0.988502  , 0.98848695, 0.98798458,\n",
       "        0.98748222, 0.98697986, 0.98647749, 0.98748277, 0.98848805,\n",
       "        0.98949333, 0.9904986 , 0.99287395, 0.9952493 , 0.99762465,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ]),\n",
       " 'unemp_insurance': 0.15,\n",
       " 'pension': 0.65,\n",
       " 'σ_ψ_init': 0.6285209184601319,\n",
       " 'init_b': 2.2162629517855197,\n",
       " 'λ': 0.0,\n",
       " 'λ_SS': 0.065,\n",
       " 'transfer': 0.0,\n",
       " 'bequest_ratio': 0.0,\n",
       " 'κ': 1.6688554994039069,\n",
       " 'σ_ψ_sub': 0.0004281045522398358,\n",
       " 'σ_θ_sub': 0.0016908499253618086,\n",
       " 'P_sub': array([[0.97569413, 0.02430587],\n",
       "        [0.02430587, 0.97569413]]),\n",
       " 'σ_ψ_2mkv': array([0.48753147, 1.39130527]),\n",
       " 'σ_θ_2mkv': array([0.0922215 , 0.26317945]),\n",
       " 'mho_2mkv': array([0.18, 0.18]),\n",
       " 'E_2mkv': array([0.96, 0.96])}"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "life_cycle_paras_q"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "## create a dictionary of parameters \n",
    "life_cycle_paras_y = {'ρ': 2.0, \n",
    "                    'β': 0.97, \n",
    "                    'P': np.array([[0.18, 0.82],\n",
    "                                   [0.04, 0.96]]), \n",
    "                    'z_val': np.array([0., 1.]), \n",
    "                    'σ_ψ': np.sqrt(0.15**2), \n",
    "                    'σ_θ': np.sqrt(0.15**2), \n",
    "                    'U': 0.0, \n",
    "                    'LivPrb': LivProbs_y,\n",
    "                    'n': pop_n,\n",
    "                    'R': 1.01, \n",
    "                    'W': 1.0, \n",
    "                    'T': T, \n",
    "                    'L': L, \n",
    "                    'G':lc_G_full, \n",
    "                    'G_sub':lc_G_full_sub, \n",
    "                    'unemp_insurance': 0.15, \n",
    "                    'pension': 0.65, \n",
    "                    'σ_ψ_init': σ_ψ_init_SCF, \n",
    "                    'init_b': b_SCF, \n",
    "                    'λ': 0.0, \n",
    "                    'λ_SS': 0.065, \n",
    "                    'transfer': 0.0, \n",
    "                    'bequest_ratio': 0.0,\n",
    "                    'κ':kappa_sipp,\n",
    "                    ## subjective profile\n",
    "                    'σ_ψ_sub':sigma_psi_sub,\n",
    "                    'σ_θ_sub':sigma_eps_sub,\n",
    "                    'P_sub': P_sub_y,\n",
    "                    'σ_ψ_2mkv':np.array([SCE_est_y.loc['$\\tilde\\sigma^l_\\psi$'],\n",
    "                                       SCE_est_y.loc['$\\tilde\\sigma^h_\\psi$']]),\n",
    "                    'σ_θ_2mkv':np.array([SCE_est_y.loc['$\\tilde\\sigma^l_\\theta$'],\n",
    "                                       SCE_est_y.loc['$\\tilde\\sigma^h_\\theta$']])\n",
    "                }\n",
    "\n",
    "\n",
    "if '$\\tilde \\mho^l$' in SCE_est_y.index:\n",
    "\n",
    "    life_cycle_paras_y['mho_2mkv'] = np.array([SCE_est_y.loc['$\\tilde \\mho^l$'],\n",
    "                                             SCE_est_y.loc['$\\tilde \\mho^h$']])\n",
    "    life_cycle_paras_y['E_2mkv'] = np.array([SCE_est_y.loc['$\\tilde E^l$'],\n",
    "                                          SCE_est_y.loc['$\\tilde E^h$']])\n",
    "    \n",
    "else:\n",
    "    life_cycle_paras_y['mho_2mkv'] = np.array([life_cycle_paras_y['P'][0,0],\n",
    "                                               life_cycle_paras_y['P'][0,0]])\n",
    "    life_cycle_paras_y['E_2mkv'] = np.array([life_cycle_paras_y['P'][1,1],\n",
    "                                           life_cycle_paras_y['P'][1,1]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ρ': 2.0,\n",
       " 'β': 0.97,\n",
       " 'P': array([[0.18, 0.82],\n",
       "        [0.04, 0.96]]),\n",
       " 'z_val': array([0., 1.]),\n",
       " 'σ_ψ': 0.15,\n",
       " 'σ_θ': 0.15,\n",
       " 'U': 0.0,\n",
       " 'LivPrb': array([0.9988675, 0.9988325, 0.9987925, 0.9987475, 0.9987015, 0.998656 ,\n",
       "        0.9986075, 0.998557 , 0.9985025, 0.9984395, 0.9983705, 0.9983025,\n",
       "        0.9982365, 0.9981685, 0.9980885, 0.9979895, 0.99787  , 0.9977265,\n",
       "        0.997558 , 0.997369 , 0.9971545, 0.9969015, 0.9966045, 0.996268 ,\n",
       "        0.9959   , 0.9955075, 0.9950905, 0.9946495, 0.994184 , 0.993683 ,\n",
       "        0.9931515, 0.9926035, 0.992042 , 0.99146  , 0.9908095, 0.990104 ,\n",
       "        0.9894025, 0.988719 , 0.988012 , 0.9872155, 0.9862755, 0.985182 ,\n",
       "        0.983905 , 0.9824375, 0.9807325, 0.978823 , 0.976777 , 0.9746125,\n",
       "        0.9722515, 0.969501 , 0.966323 , 0.9628335, 0.9590265, 0.95479  ,\n",
       "        0.9498745, 0.9442255, 0.93797  , 0.93109  , 0.9234355, 0.914817 ]),\n",
       " 'n': 0.005,\n",
       " 'R': 1.01,\n",
       " 'W': 1.0,\n",
       " 'T': 40,\n",
       " 'L': 60,\n",
       " 'G': array([1.06609449, 1.06191502, 1.05791078, 1.05407258, 1.05039145,\n",
       "        1.04685857, 1.04346528, 1.0402031 , 1.03706367, 1.03403875,\n",
       "        1.03112026, 1.02830021, 1.02557071, 1.02292399, 1.02035237,\n",
       "        1.01784826, 1.01540414, 1.01301259, 1.01066625, 1.00835783,\n",
       "        1.00608013, 1.00382599, 1.00158834, 0.99936014, 0.99713445,\n",
       "        0.99490436, 0.99266305, 0.99040373, 0.98811971, 0.98580433,\n",
       "        0.98345102, 0.98105328, 0.97860466, 0.97609881, 0.97352944,\n",
       "        0.97089036, 0.96817545, 0.96537869, 0.96249415, 0.95951603,\n",
       "        0.70770167, 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ]),\n",
       " 'G_sub': array([1.00601101, 1.0019325 , 1.00299144, 0.99931616, 0.99995637,\n",
       "        1.00123227, 0.99675304, 0.99811018, 0.99805862, 0.99481797,\n",
       "        0.99882144, 0.99544615, 0.98350757, 0.99665678, 0.99394703,\n",
       "        0.99421018, 0.99365354, 0.99338818, 0.99405426, 0.99376881,\n",
       "        0.99240822, 0.99057603, 0.99004096, 0.99303108, 0.96195108,\n",
       "        0.9886356 , 0.98934692, 0.98638594, 0.98828399, 0.98677778,\n",
       "        0.9915061 , 0.98730075, 0.98665285, 0.98855507, 0.99132812,\n",
       "        0.98479033, 0.98854715, 0.98848695, 0.98647749, 0.9904986 ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ,\n",
       "        1.        , 1.        , 1.        , 1.        , 1.        ]),\n",
       " 'unemp_insurance': 0.15,\n",
       " 'pension': 0.65,\n",
       " 'σ_ψ_init': 0.6285209184601319,\n",
       " 'init_b': 0.5540657379463798,\n",
       " 'λ': 0.0,\n",
       " 'λ_SS': 0.065,\n",
       " 'transfer': 0.0,\n",
       " 'bequest_ratio': 0.0,\n",
       " 'κ': 1.6688554994039069,\n",
       " 'σ_ψ_sub': 0.034311623666908396,\n",
       " 'σ_θ_sub': 0.02055997279522646,\n",
       " 'P_sub': array([[0.97377031, 0.02622969],\n",
       "        [0.02622969, 0.97377031]]),\n",
       " 'σ_ψ_2mkv': array([0.97448242, 2.66183019]),\n",
       " 'σ_θ_2mkv': array([0.02304165, 0.06293901]),\n",
       " 'mho_2mkv': array([0.18, 0.18]),\n",
       " 'E_2mkv': array([0.96, 0.96])}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "life_cycle_paras_y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Production function parameters "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "production_paras_y={}\n",
    "production_paras_y['K2Y ratio'] = 3.0\n",
    "production_paras_y['W'] = 1.0\n",
    "production_paras_y['α'] = 0.33\n",
    "production_paras_y['δ'] = 0.025"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "## quarterly paras \n",
    "production_paras_q = copy(production_paras_y)\n",
    "production_paras_q['K2Y ratio'] = production_paras_q['K2Y ratio']*4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "lines_to_next_cell": 2
   },
   "source": [
    "### Export the parameters into a table used in the draft"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "life_cycle_paras_y_copy = copy(life_cycle_paras_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "del life_cycle_paras_y_copy['G']  \n",
    "del life_cycle_paras_y_copy['G_sub']  \n",
    "del life_cycle_paras_y_copy['σ_ψ_2mkv']\n",
    "del life_cycle_paras_y_copy['σ_θ_2mkv']\n",
    "\n",
    "if 'mho_2mkv' in life_cycle_paras_y_copy.keys():\n",
    "    del life_cycle_paras_y_copy['mho_2mkv']  \n",
    "if 'E_2mkv' in life_cycle_paras_y_copy.keys():\n",
    "    del life_cycle_paras_y_copy['E_2mkv']  \n",
    "del life_cycle_paras_y_copy['P']  \n",
    "del life_cycle_paras_y_copy['z_val']  \n",
    "del life_cycle_paras_y_copy['U']  \n",
    "del life_cycle_paras_y_copy['κ']  \n",
    "del life_cycle_paras_y_copy['P_sub']  \n",
    "#del life_cycle_paras_y_copy['q']  \n",
    "#del life_cycle_paras_y_copy['p']  \n",
    "del life_cycle_paras_y_copy['transfer']  \n",
    "\n",
    "\n",
    "## rewrite some parameters' names\n",
    "\n",
    "life_cycle_paras_y_copy['U2U'] = life_cycle_paras_y['P'][0,0]\n",
    "life_cycle_paras_y_copy['E2E'] = life_cycle_paras_y['P'][1,1]\n",
    "\n",
    "## rename some \n",
    "#life_cycle_paras_y_copy['1-D'] =  life_cycle_paras_y_copy.pop('LivPrb')\n",
    "life_cycle_paras_y_copy['μ'] =  life_cycle_paras_y_copy.pop('unemp_insurance')\n",
    "life_cycle_paras_y_copy['b_init'] =  life_cycle_paras_y_copy.pop('init_b')\n",
    "\n",
    "## rounding \n",
    "#life_cycle_paras_y_copy['1-D'] = round(life_cycle_paras_y_copy['1-D'],3)\n",
    "life_cycle_paras_y_copy['σ_ψ_init'] = round(life_cycle_paras_y_copy['σ_ψ_init'],3)\n",
    "life_cycle_paras_y_copy['b_init'] = round(life_cycle_paras_y_copy['b_init'],3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "## merge life-cycle and production paras into model paras\n",
    "\n",
    "model_paras = copy(life_cycle_paras_y_copy)\n",
    "model_paras.update(production_paras_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ρ': 2.0,\n",
       " 'β': 0.97,\n",
       " 'σ_ψ': 0.15,\n",
       " 'σ_θ': 0.15,\n",
       " 'LivPrb': array([0.9988675, 0.9988325, 0.9987925, 0.9987475, 0.9987015, 0.998656 ,\n",
       "        0.9986075, 0.998557 , 0.9985025, 0.9984395, 0.9983705, 0.9983025,\n",
       "        0.9982365, 0.9981685, 0.9980885, 0.9979895, 0.99787  , 0.9977265,\n",
       "        0.997558 , 0.997369 , 0.9971545, 0.9969015, 0.9966045, 0.996268 ,\n",
       "        0.9959   , 0.9955075, 0.9950905, 0.9946495, 0.994184 , 0.993683 ,\n",
       "        0.9931515, 0.9926035, 0.992042 , 0.99146  , 0.9908095, 0.990104 ,\n",
       "        0.9894025, 0.988719 , 0.988012 , 0.9872155, 0.9862755, 0.985182 ,\n",
       "        0.983905 , 0.9824375, 0.9807325, 0.978823 , 0.976777 , 0.9746125,\n",
       "        0.9722515, 0.969501 , 0.966323 , 0.9628335, 0.9590265, 0.95479  ,\n",
       "        0.9498745, 0.9442255, 0.93797  , 0.93109  , 0.9234355, 0.914817 ]),\n",
       " 'n': 0.005,\n",
       " 'R': 1.01,\n",
       " 'W': 1.0,\n",
       " 'T': 40,\n",
       " 'L': 60,\n",
       " 'pension': 0.65,\n",
       " 'σ_ψ_init': 0.629,\n",
       " 'λ': 0.0,\n",
       " 'λ_SS': 0.065,\n",
       " 'bequest_ratio': 0.0,\n",
       " 'σ_ψ_sub': 0.034311623666908396,\n",
       " 'σ_θ_sub': 0.02055997279522646,\n",
       " 'U2U': 0.18,\n",
       " 'E2E': 0.96,\n",
       " 'μ': 0.15,\n",
       " 'b_init': 0.554,\n",
       " 'K2Y ratio': 3.0,\n",
       " 'α': 0.33,\n",
       " 'δ': 0.025}"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_paras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "# making blocks \n",
    "\n",
    "blocknames =['risk',\n",
    "            'initial condition',\n",
    "            'life cycle',\n",
    "            'preference',\n",
    "            'policy',\n",
    "            'production',\n",
    "            'subjective']\n",
    "\n",
    "prefernece = ['ρ','β']\n",
    "lifecycle = ['T','L','n',\n",
    "             #'1-D'\n",
    "            ]\n",
    "risk  = ['σ_ψ','σ_θ','U2U','E2E']\n",
    "initial = ['σ_ψ_init','init_b','bequest_ratio']\n",
    "policy = ['μ','pension','λ','λ_SS']\n",
    "production=['K2Y ratio','W','α','δ']\n",
    "subjective= ['σ_ψ_sub','σ_θ_sub']\n",
    "\n",
    "block_all= [risk,\n",
    "            initial,\n",
    "            lifecycle,\n",
    "            prefernece,\n",
    "            policy,\n",
    "           production,\n",
    "           subjective]\n",
    "\n",
    "## create multiple layer dictionary \n",
    "\n",
    "model_paras_by_block = {}\n",
    "\n",
    "for i,block in enumerate(block_all):\n",
    "    model_paras_by_block[blocknames[i]] =  {k: v for k, v \n",
    "                                               in model_paras.items() \n",
    "                                               if k in block}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'risk': {'σ_ψ': 0.15, 'σ_θ': 0.15, 'U2U': 0.18, 'E2E': 0.96},\n",
       " 'initial condition': {'σ_ψ_init': 0.629, 'bequest_ratio': 0.0},\n",
       " 'life cycle': {'n': 0.005, 'T': 40, 'L': 60},\n",
       " 'preference': {'ρ': 2.0, 'β': 0.97},\n",
       " 'policy': {'pension': 0.65, 'λ': 0.0, 'λ_SS': 0.065, 'μ': 0.15},\n",
       " 'production': {'W': 1.0, 'K2Y ratio': 3.0, 'α': 0.33, 'δ': 0.025},\n",
       " 'subjective': {'σ_ψ_sub': 0.034311623666908396,\n",
       "  'σ_θ_sub': 0.02055997279522646}}"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_paras_by_block"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_paras_by_block_df = pd.DataFrame.from_dict({(i,j): model_paras_by_block[i][j] \n",
    "                           for i in model_paras_by_block.keys() \n",
    "                           for j in model_paras_by_block[i].keys()},\n",
    "                       orient='index')\n",
    "\n",
    "model_paras_by_block_df.columns =['values']\n",
    "Mindex = pd.MultiIndex.from_tuples(list(model_paras_by_block_df.index), names=[\"block\", \"parameter\"])\n",
    "model_paras_by_block_df.index = Mindex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_paras_by_block_df['source']=''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>source</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>block</th>\n",
       "      <th>parameter</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">risk</th>\n",
       "      <th>σ_ψ</th>\n",
       "      <td>0.150000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>σ_θ</th>\n",
       "      <td>0.150000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>U2U</th>\n",
       "      <td>0.180000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>E2E</th>\n",
       "      <td>0.960000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">initial condition</th>\n",
       "      <th>σ_ψ_init</th>\n",
       "      <td>0.629000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>bequest_ratio</th>\n",
       "      <td>0.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">life cycle</th>\n",
       "      <th>n</th>\n",
       "      <td>0.005000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>T</th>\n",
       "      <td>40.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>L</th>\n",
       "      <td>60.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">preference</th>\n",
       "      <th>ρ</th>\n",
       "      <td>2.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>β</th>\n",
       "      <td>0.970000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">policy</th>\n",
       "      <th>pension</th>\n",
       "      <td>0.650000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>λ</th>\n",
       "      <td>0.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>λ_SS</th>\n",
       "      <td>0.065000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>μ</th>\n",
       "      <td>0.150000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">production</th>\n",
       "      <th>W</th>\n",
       "      <td>1.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K2Y ratio</th>\n",
       "      <td>3.000000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>α</th>\n",
       "      <td>0.330000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>δ</th>\n",
       "      <td>0.025000</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">subjective</th>\n",
       "      <th>σ_ψ_sub</th>\n",
       "      <td>0.034312</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>σ_θ_sub</th>\n",
       "      <td>0.020560</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    values source\n",
       "block             parameter                      \n",
       "risk              σ_ψ             0.150000       \n",
       "                  σ_θ             0.150000       \n",
       "                  U2U             0.180000       \n",
       "                  E2E             0.960000       \n",
       "initial condition σ_ψ_init        0.629000       \n",
       "                  bequest_ratio   0.000000       \n",
       "life cycle        n               0.005000       \n",
       "                  T              40.000000       \n",
       "                  L              60.000000       \n",
       "preference        ρ               2.000000       \n",
       "                  β               0.970000       \n",
       "policy            pension         0.650000       \n",
       "                  λ               0.000000       \n",
       "                  λ_SS            0.065000       \n",
       "                  μ               0.150000       \n",
       "production        W               1.000000       \n",
       "                  K2Y ratio       3.000000       \n",
       "                  α               0.330000       \n",
       "                  δ               0.025000       \n",
       "subjective        σ_ψ_sub         0.034312       \n",
       "                  σ_θ_sub         0.020560       "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_paras_by_block_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [],
   "source": [
    "## add source of the parameters \n",
    "\n",
    "model_paras_by_block_df.loc['risk','source']='median estimates from the literature'\n",
    "model_paras_by_block_df.loc['initial condition','source']='estimated for age 25 in the 2016 SCF'\n",
    "model_paras_by_block_df.loc[('initial condition','bequest_ratio'),'source']='assumption'\n",
    "model_paras_by_block_df.loc[('life cycle','T'),'source']='standard calibration'\n",
    "model_paras_by_block_df.loc[('life cycle','L'),'source']='standard calibration'\n",
    "model_paras_by_block_df.loc[('life cycle','n'),'source']='U.S. Census'\n",
    "model_paras_by_block_df.loc[('preference','β'),'source']='targeting average wealth/income ratio'\n",
    "model_paras_by_block_df.loc[('preference','ρ'),'source']='standard calibration'\n",
    "model_paras_by_block_df.loc['policy','source']='U.S. average'\n",
    "model_paras_by_block_df.loc[('policy','λ'),'values']=np.nan\n",
    "model_paras_by_block_df.loc[('policy','λ'),'source']='endogenously determined'\n",
    "model_paras_by_block_df.loc[('policy','λ_SS'),'source']='U.S. average'\n",
    "model_paras_by_block_df.loc[('policy','λ_SS'),'values']=np.nan\n",
    "\n",
    "model_paras_by_block_df.loc['production','source']='standard calibration'\n",
    "model_paras_by_block_df.loc[('production','W'),'source']='target values in steady state'\n",
    "model_paras_by_block_df.loc[('production','K2Y ratio'),'source']='target values in steady state'\n",
    "model_paras_by_block_df.loc['subjective','source']='estimated from SCE'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "## for latex symbols \n",
    "parameter_list = [para[1] for para in list(model_paras_by_block_df.index)]\n",
    "\n",
    "para_latex = ['$\\\\sigma_\\\\psi$',\n",
    "              '$\\\\sigma_\\\\theta$',\n",
    "              '$U2U$',\n",
    "              '$E2E$',\n",
    "              '$\\\\sigma_\\\\psi^{\\\\text{init}}$',\n",
    "              'bequest ratio',\n",
    "               '$n$',\n",
    "              '$T$',\n",
    "              '$L$',        \n",
    "              #'$1-D$',\n",
    "              '$\\\\rho$',\n",
    "              '$\\\\beta$',\n",
    "              '$\\\\mathbb{S}$',\n",
    "             '$\\\\lambda$',\n",
    "             '$\\\\lambda_{SS}$',\n",
    "             '$\\\\mu$',\n",
    "             '$W$',\n",
    "             'K2Y ratio',\n",
    "             '$\\\\alpha$',\n",
    "             '$\\\\delta$',\n",
    "              '$\\\\sigma_\\\\psi^{\\\\text{sub}}$',\n",
    "             '$\\\\sigma_\\\\theta^{\\\\text{sub}}$']\n",
    "\n",
    "model_paras_by_block_df['parameter name']= para_latex\n",
    "\n",
    "model_paras_by_block_df = model_paras_by_block_df[['parameter name','values','source']]\n",
    "\n",
    "model_paras_by_block_df=model_paras_by_block_df.reset_index(level=1, drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "## export to excel \n",
    "model_paras_by_block_df.to_excel('../Tables/calibration.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>parameter name</th>\n",
       "      <th>values</th>\n",
       "      <th>source</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>block</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>risk</th>\n",
       "      <td>$\\sigma_\\psi$</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>median estimates from the literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>risk</th>\n",
       "      <td>$\\sigma_\\theta$</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>median estimates from the literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>risk</th>\n",
       "      <td>$U2U$</td>\n",
       "      <td>0.180000</td>\n",
       "      <td>median estimates from the literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>risk</th>\n",
       "      <td>$E2E$</td>\n",
       "      <td>0.960000</td>\n",
       "      <td>median estimates from the literature</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>initial condition</th>\n",
       "      <td>$\\sigma_\\psi^{\\text{init}}$</td>\n",
       "      <td>0.629000</td>\n",
       "      <td>estimated for age 25 in the 2016 SCF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>initial condition</th>\n",
       "      <td>bequest ratio</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>assumption</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>life cycle</th>\n",
       "      <td>$n$</td>\n",
       "      <td>0.005000</td>\n",
       "      <td>U.S. Census</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>life cycle</th>\n",
       "      <td>$T$</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>standard calibration</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>life cycle</th>\n",
       "      <td>$L$</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>standard calibration</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>preference</th>\n",
       "      <td>$\\rho$</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>standard calibration</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>preference</th>\n",
       "      <td>$\\beta$</td>\n",
       "      <td>0.970000</td>\n",
       "      <td>calibrated to match average wealth/income ratio</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy</th>\n",
       "      <td>$\\mathbb{S}$</td>\n",
       "      <td>0.650000</td>\n",
       "      <td>U.S. average</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy</th>\n",
       "      <td>$\\lambda$</td>\n",
       "      <td>NaN</td>\n",
       "      <td>endogenously determined</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy</th>\n",
       "      <td>$\\lambda_{SS}$</td>\n",
       "      <td>NaN</td>\n",
       "      <td>U.S. average</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>policy</th>\n",
       "      <td>$\\mu$</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>U.S. average</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>production</th>\n",
       "      <td>$W$</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>target values in steady state</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>production</th>\n",
       "      <td>K2Y ratio</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>target values in steady state</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>production</th>\n",
       "      <td>$\\alpha$</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>standard calibration</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>production</th>\n",
       "      <td>$\\delta$</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>standard calibration</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subjective</th>\n",
       "      <td>$\\sigma_\\psi^{\\text{sub}}$</td>\n",
       "      <td>0.034312</td>\n",
       "      <td>estimated from SCE</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subjective</th>\n",
       "      <td>$\\sigma_\\theta^{\\text{sub}}$</td>\n",
       "      <td>0.020560</td>\n",
       "      <td>estimated from SCE</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 parameter name     values  \\\n",
       "block                                                        \n",
       "risk                              $\\sigma_\\psi$   0.150000   \n",
       "risk                            $\\sigma_\\theta$   0.150000   \n",
       "risk                                      $U2U$   0.180000   \n",
       "risk                                      $E2E$   0.960000   \n",
       "initial condition   $\\sigma_\\psi^{\\text{init}}$   0.629000   \n",
       "initial condition                 bequest ratio   0.000000   \n",
       "life cycle                                  $n$   0.005000   \n",
       "life cycle                                  $T$  40.000000   \n",
       "life cycle                                  $L$  60.000000   \n",
       "preference                               $\\rho$   2.000000   \n",
       "preference                              $\\beta$   0.970000   \n",
       "policy                             $\\mathbb{S}$   0.650000   \n",
       "policy                                $\\lambda$        NaN   \n",
       "policy                           $\\lambda_{SS}$        NaN   \n",
       "policy                                    $\\mu$   0.150000   \n",
       "production                                  $W$   1.000000   \n",
       "production                            K2Y ratio   3.000000   \n",
       "production                             $\\alpha$   0.330000   \n",
       "production                             $\\delta$   0.025000   \n",
       "subjective           $\\sigma_\\psi^{\\text{sub}}$   0.034312   \n",
       "subjective         $\\sigma_\\theta^{\\text{sub}}$   0.020560   \n",
       "\n",
       "                                                            source  \n",
       "block                                                               \n",
       "risk                          median estimates from the literature  \n",
       "risk                          median estimates from the literature  \n",
       "risk                          median estimates from the literature  \n",
       "risk                          median estimates from the literature  \n",
       "initial condition             estimated for age 25 in the 2016 SCF  \n",
       "initial condition                                       assumption  \n",
       "life cycle                                             U.S. Census  \n",
       "life cycle                                    standard calibration  \n",
       "life cycle                                    standard calibration  \n",
       "preference                                    standard calibration  \n",
       "preference         calibrated to match average wealth/income ratio  \n",
       "policy                                                U.S. average  \n",
       "policy                                     endogenously determined  \n",
       "policy                                                U.S. average  \n",
       "policy                                                U.S. average  \n",
       "production                           target values in steady state  \n",
       "production                           target values in steady state  \n",
       "production                                    standard calibration  \n",
       "production                                    standard calibration  \n",
       "subjective                                      estimated from SCE  \n",
       "subjective                                      estimated from SCE  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_paras_by_block_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "jupytext": {
   "encoding": "# -*- coding: utf-8 -*-",
   "formats": "ipynb,py:light"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}