Skip to content

Commit

Permalink
Fixes #31 - documentation
Browse files Browse the repository at this point in the history
  • Loading branch information
@bjornerstedt
bjornerstedt committed Jan 13, 2016
1 parent 9698535 commit 66b39b2
Show file tree
Hide file tree
Showing 7 changed files with 708 additions and 579 deletions.
1 change: 1 addition & 0 deletions .gitignore
View file
Original file line number Diff line number Diff line change
Expand Up @@ -7,3 +7,4 @@
*.dta
*.pdf
*.m~
*.lyx#
210 changes: 107 additions & 103 deletions trunk/doc/reference_doc.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
<!--
This HTML was auto-generated from MATLAB code.
To make changes, update the MATLAB code and republish this document.
--><title>SimMarket 0.2 Reference</title><meta name="generator" content="MATLAB 8.6"><link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"><meta name="DC.date" content="2016-01-10"><meta name="DC.source" content="reference_doc.m"><style type="text/css">
--><title>SimMarket 0.2 Reference</title><meta name="generator" content="MATLAB 8.6"><link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"><meta name="DC.date" content="2016-01-13"><meta name="DC.source" content="reference_doc.m"><style type="text/css">
html,body,div,span,applet,object,iframe,h1,h2,h3,h4,h5,h6,p,blockquote,pre,a,abbr,acronym,address,big,cite,code,del,dfn,em,font,img,ins,kbd,q,s,samp,small,strike,strong,sub,sup,tt,var,b,u,i,center,dl,dt,dd,ol,ul,li,fieldset,form,label,legend,table,caption,tbody,tfoot,thead,tr,th,td{margin:0;padding:0;border:0;outline:0;font-size:100%;vertical-align:baseline;background:transparent}body{line-height:1}ol,ul{list-style:none}blockquote,q{quotes:none}blockquote:before,blockquote:after,q:before,q:after{content:'';content:none}:focus{outine:0}ins{text-decoration:none}del{text-decoration:line-through}table{border-collapse:collapse;border-spacing:0}

html { min-height:100%; margin-bottom:1px; }
Expand Down Expand Up @@ -79,7 +79,7 @@

Estimate with properties:

data: [500x17 table]
data: [500x11 table]
panelid: []
results: [1x1 struct]
y: []
Expand All @@ -103,12 +103,12 @@

SettingsClass with properties:

market: []
panel: []
exog: []
instruments: []
endog: []
exog: []
panel: []
depvar: []
market: []

</pre><p>The est.settings structure has the following fields</p><pre> robust: 1 - robust estimation true/false
paneltype: 'none' - panel estimate: 'fe'/'lsdv'/'none'
Expand All @@ -119,10 +119,10 @@

SettingsClass with properties:

paneltype: 'fe'
robust: 1
estimateMethod: 'ols'
nocons: 0
paneltype: 'fe'

</pre><p>The method <tt>Estimate.estimate()</tt> generates a result table as output. It also populates the structure <tt>Estimate.results</tt> with various results</p><p>estimateDescription: 'Linear Estimate' other: [x] params: Structure with the estimate and var-covar matrix estimate: Estimate table var: Table with variable names used settings: Table with settings</p><pre class="codeinput">est.settings.paneltype = <span class="string">'none'</span>;
est.var.exog = <span class="string">'w'</span>;
Expand Down Expand Up @@ -155,23 +155,23 @@
names values
_____________ ______

'market' []
'panel' []
'exog' 'w'
'instruments' []
'endog' []
'exog' 'w'
'panel' []
'depvar' 'c'
'market' []


ans =

names values
________________ ______

'paneltype' 'none'
'robust' [ 1]
'estimateMethod' 'ols'
'nocons' [ 0]
'paneltype' 'none'

</pre><p>In estimating, Estimate creates the following properties that can be used in calculations. To generate these properties without estimating, the method <tt>Estimate.init()</tt> is invoked.</p><pre> y: []
X: []
Expand Down Expand Up @@ -199,7 +199,7 @@
sigma: []
d: []
marketid: []
data: [500x9 table]
data: [500x8 table]
panelid: []
results: [1x1 struct]
y: []
Expand All @@ -217,16 +217,16 @@

SettingsClass with properties:

market: []
nests: []
instruments: []
quantity: []
marketsize: []
nests: []
price: []
endog: []
exog: []
panel: []
exog: []
instruments: []
endog: []
depvar: []
market: []
marketsize: []

</pre><p>NLDemand has the following additional properties:</p><pre> alpha: The calibrated or estimated alpha parameter
sigma: A vector with sigmas
Expand All @@ -239,11 +239,11 @@

SettingsClass with properties:

paneltype: 'lsdv'
robust: 1
estimateMethod: 'gls'
ces: 0
estimateMethod: 'gls'
nocons: 0
paneltype: 'fe'

</pre><p><b>Methods</b></p><p>The method <tt>NLDemand.estimate()</tt> performs a linear panel estimate based on the settings.</p><pre class="codeinput">methods(NLDemand)
</pre><pre class="codeoutput">
Expand All @@ -267,7 +267,7 @@
sigma: []
d: []
marketid: []
data: [500x9 table]
data: [500x8 table]
panelid: []
results: [1x1 struct]
y: []
Expand All @@ -285,19 +285,19 @@

SettingsClass with properties:

market: []
nests: []
instruments: []
quantity: []
marketsize: []
nests: []
price: []
nonlinearlogs: []
endog: []
exog: []
nonlinear: []
panel: []
nonlineartriangular: []
panel: []
nonlinear: []
exog: []
instruments: []
endog: []
depvar: []
market: []
marketsize: []

</pre><p>RCDemand with properties:</p><pre> sigma: The calibrated or estimated nonlinear parameters</pre><p>Settings</p><p>RCDemand.settings has properties:</p><pre> ces: 0 - CES or Unit logit demand
maxiter: 100 - Maximum number of iterations in optimization
Expand All @@ -314,18 +314,18 @@

SettingsClass with properties:

maxiter: 100
robust: 1
estimateMethod: 'gls'
quaddraws: 10
paneltype: 'lsdv'
sigma0: []
nind: 100
optimalIV: 0
robust: 1
maxiter: 100
drawmethod: 'hypercube'
ces: 0
optimalIV: 0
marketdraws: 0
sigma0: []
quaddraws: 10
estimateMethod: 'gls'
nocons: 0
paneltype: 'lsdv'
drawmethod: 'hypercube'

</pre><p>RCDemand.config</p><pre> hessian: 0
test: []
Expand All @@ -341,17 +341,17 @@

SettingsClass with properties:

fpmaxit: 1000
randstream: []
test: []
quietly: 1
hessian: 0
tolerance: 1.0000e-09
fpmaxit: 1000
guessdelta: 1
fptolerance2: 1.0000e-14
restartFval: 1000
restartMaxIterations: 1
test: []
fptolerance2: 1.0000e-14
fptolerance1: 1.0000e-14
quietly: 1
randstream: []
guessdelta: 1
hessian: 0

</pre><p><b>Methods</b></p><p>The method <tt>RCDemand.estimate()</tt> performs a BLP estimate based on the settings specified in the demand object.</p><pre class="codeinput">methods(RCDemand)
</pre><pre class="codeoutput">
Expand All @@ -375,13 +375,13 @@

SettingsClass with properties:

market: []
firm: []
panel: []
exog: []
instruments: []
endog: []
exog: []
panel: []
depvar: []
market: []
firm: []

</pre><p><tt>Market.var.firm</tt> is the only property that has to be set to calculate costs or equilibrium.</p><pre class="codeinput">market.var.firm = <span class="string">'productid'</span>;
</pre><p>An estimated or calibrated demand is associated with the marktet either by providing it in the constructor or adding it to the <tt>Market.demand</tt> property:</p><pre class="codeinput">market.demand = demand;
Expand All @@ -395,15 +395,15 @@

SettingsClass with properties:

conduct: 0
paneltype: 'fe'
maxit: 3000
dampen: 1
robust: 1
weightedAverages: 1
estimateMethod: 'ols'
conduct: 0
robust: 1
valueShares: 0
estimateMethod: 'ols'
nocons: 0
paneltype: 'fe'
maxit: 3000

</pre><p><b>Methods</b></p><p><tt>Market.findCosts()</tt> calculates costs based on a demand specification Prices and quantities used are copied from the demand specification</p><pre class="language-matlab">market.findCosts()
</pre><p><tt>Market.equilibrium()</tt> calculates a market equilibrium based on a demand specification, costs, and a specification of ownership and conduct (using <tt>Market.var.firm</tt> and <tt>Market.settings.conduct</tt>.</p><pre class="codeinput">methods(Market)
Expand Down Expand Up @@ -435,54 +435,58 @@
demand: []
market: []

</pre><p>The market model is specified with the parameters in <tt>SimMarket.model</tt>:</p><pre> endog: 0 - Endogenous prices and quanities true/false
randproducts: 0 - Exogenously random products in market true/false
simulatePrices: 1 - Simulate prices or let them be randomly drawn as
</pre><p>The market model is specified with the parameters in <tt>SimMarket.model</tt>:</p><pre> endog: 0 - Endogenous prices and quanities true/false
randomProducts: 0 - Exogenously random products in market true/false
pricesFromCosts: 1 - Simulate prices or let them be randomly drawn as
in Nevo code true/false
markets: 100 - Number of markets generated
products: 5 - (Maximum) number of products in each market.
types: [] - Number of types for each categorical
firm: [] - Vector of ownership for each producty
beta: [1 0] -
x: [5 0] - Expected value for p and other demand shifters
x_vcv: [1 1] - Variance, can be specified as a matrix for
markets: 100 - Number of markets generated
products: 5 - (Maximum) number of products in each market.
types: [] - Number of types for each categorical
firm: [] - Vector of ownership for each producty
beta: [1 0] -
x: [5 0] - Expected value for p and other demand shifters
x_vcv: [1 1] - Variance, can be specified as a matrix for
multicollinearity
c: 4 - Costs
c_vcv: 1
gamma: 0 - Cost shifter parameter
epsilon_sigma: 0.1 - Sd of individual unobservables
sigma_xi: 0.1 - Sd of panel unobservables
endog_sigma: 0.1 - Endogeneity parameter for non simulated prices
prob_prod: 0.8 - Probability that product exists in a market</pre><pre class="codeinput">m.model
c: 4 - Costs
c_vcv: 1
gamma: 0 - Cost shifter parameter
epsilon_sigma: 0.1 - Sd of individual unobservables
sigma_xi: 0.1 - Sd of panel unobservables
endog_sigma: 0.1 - Endogeneity parameter for non simulated prices
prob_prod: 0.8 - Probability that product exists in a market</pre><pre class="codeinput">m.model
</pre><pre class="codeoutput">
ans =

SettingsClass with properties:

types: []
epsilon_sigma: 0.1000
x: [5 0]
x_vcv: [1 1]
c_vcv: 1
firm: []
sigma_xi: 0.1000
beta: [1 0]
endog: 0
prob_prod: 0.8000
endog_sigma: 0.1000
c: 4
simulatePrices: 1
products: 5
gamma: 0
markets: 100
randproducts: 0
products: 5
pricesFromCosts: 1
beta: [1 0]
sigma_xi: 0.1000
prob_prod: 0.8000
x_vcv: [1 1]
markets: 100
x: [5 0]
c_vcv: 1
endog: 0
randomProducts: 0
epsilon_sigma: 0.1000
gamma: 0
endog_sigma: 0.1000
c: 4
types: []
firm: []

</pre><p><b>Methods</b></p><div><ul><li>SimMarket - Create a new simulation object, optionally with demand spec</li><li>create - Creates market - should return dataset.</li><li>estimate - Estimate and compare, used in testing framework</li><li>findCosts - Calculate costs, used in testing framework</li></ul></div><pre class="codeinput">methods(SimMarket)
</pre><pre class="codeoutput">
Methods for class SimMarket:

SimMarket copy create estimate findCosts means

Static methods:

testEqual

Call "methods('handle')" for methods of SimMarket inherited from handle.

</pre><p class="footer"><br><a href="http://www.mathworks.com/products/matlab/">Published with MATLAB&reg; R2015b</a><br></p></div><!--
Expand Down Expand Up @@ -751,25 +755,25 @@
%%
% The market model is specified with the parameters in |SimMarket.model|:
%
% endog: 0 - Endogenous prices and quanities true/false
% randproducts: 0 - Exogenously random products in market true/false
% simulatePrices: 1 - Simulate prices or let them be randomly drawn as
% endog: 0 - Endogenous prices and quanities true/false
% randomProducts: 0 - Exogenously random products in market true/false
% pricesFromCosts: 1 - Simulate prices or let them be randomly drawn as
% in Nevo code true/false
% markets: 100 - Number of markets generated
% products: 5 - (Maximum) number of products in each market.
% types: [] - Number of types for each categorical
% firm: [] - Vector of ownership for each producty
% beta: [1 0] -
% x: [5 0] - Expected value for p and other demand shifters
% x_vcv: [1 1] - Variance, can be specified as a matrix for
% markets: 100 - Number of markets generated
% products: 5 - (Maximum) number of products in each market.
% types: [] - Number of types for each categorical
% firm: [] - Vector of ownership for each producty
% beta: [1 0] -
% x: [5 0] - Expected value for p and other demand shifters
% x_vcv: [1 1] - Variance, can be specified as a matrix for
% multicollinearity
% c: 4 - Costs
% c_vcv: 1
% gamma: 0 - Cost shifter parameter
% epsilon_sigma: 0.1 - Sd of individual unobservables
% sigma_xi: 0.1 - Sd of panel unobservables
% endog_sigma: 0.1 - Endogeneity parameter for non simulated prices
% prob_prod: 0.8 - Probability that product exists in a market
% c: 4 - Costs
% c_vcv: 1
% gamma: 0 - Cost shifter parameter
% epsilon_sigma: 0.1 - Sd of individual unobservables
% sigma_xi: 0.1 - Sd of panel unobservables
% endog_sigma: 0.1 - Endogeneity parameter for non simulated prices
% prob_prod: 0.8 - Probability that product exists in a market
m.model
%%
Expand Down
Loading

0 comments on commit 66b39b2

Please sign in to comment.