-
Notifications
You must be signed in to change notification settings - Fork 9
/
Panel1.html
309 lines (239 loc) · 9.98 KB
/
Panel1.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
<!DOCTYPE html>
<html lang="" xml:lang="">
<head>
<title>Panal Data 1: Framework</title>
<meta charset="utf-8" />
<meta name="author" content="Instructor: Yuta Toyama" />
<script src="Panel1_files/header-attrs-2.6/header-attrs.js"></script>
<link rel="stylesheet" href="xaringan-themer.css" type="text/css" />
</head>
<body>
<textarea id="source">
class: center, middle, inverse, title-slide
# Panal Data 1: Framework
### Instructor: Yuta Toyama
### Last updated: 2021-07-09
---
class: title-slide-section, center, middle
name: logistics
# Introduction
---
## Introduction
- Panel data (パネルデータ)
- combination of crosssection (クロスセクション) and time series (時系列) data
- Examples:
1. Person `\(i\)`'s income in year `\(t\)`.
2. Vote share in county `\(i\)` for the presidential election year `\(t\)`.
3. Country `\(i\)`'s GDP in year `\(t\)`.
- Panel data is useful
1. More variation (both cross-sectional and time series variation)
2. Can deal with **time-invariant unobserved factors**.
---
## Course Plan
- Framework
- Implementation in R
- **Difference-in-differences (DID, 差の差分法)**
---
class: title-slide-section, center, middle
name: logistics
# Framework
---
## Framework with Panel Data
- Consider the model
`$$y_{it} = \beta' x_{it} + \epsilon_{it}, E[\epsilon_{it} | x_{it} ] = 0$$`
where `\(x_{it}\)` is a k-dimensional vector
- If there is no correlation between `\(x_{it}\)` and `\(\epsilon_{it}\)`, you can estimate the model by OLS **(pooled OLS)**
- A concern here is the omitted variable bias.
---
## Introducing fixed effect (固定効果)
- Suppose that `\(\epsilon_{it}\)` is decomposed as
`$$\epsilon_{it} = \alpha_i + u_{it}$$`
where `\(\alpha_i\)` is called **unit fixed effect (固定効果)**, which is the time-invariant unobserved heterogeneity.
- With panel data, we can control for the unit fixed effects by incorporating the dummy variable for each unit `\(i\)`!
`$$y_{it} = \beta' x_{it} + \gamma_2 D2_i + \cdots + \gamma_n Dn_i + u_{it}$$`
where `\(Dl_i\)` takes 1 if `\(l=i\)`.
---
## Fixed Effect Model
- Model
`$$y_{it} = \beta' x_{it} + \alpha_i + u_{it}$$`
- Assumptions:
1. `\(u_{it}\)` is uncorrelated with `\((x_{i1},\cdots, x_{iT})\)`, that is `\(E[u_{it}|x_{i1},\cdots, x_{iT} ] = 0\)`
2. `\((Y_{it}, x_{it})\)` are independent across individual `\(i\)`.
3. No outliers
4. No perfect multicollinarity between explantory variables `\(x_{it}\)` and fixed effects `\(\alpha_i\)`.
---
## Assumption 1: Mean independence
- Assumption 1 is weaker than the assumption in OLS.
- Here, the time-invariant unobserved factor is captured by the fixed effect `\(\alpha_i\)`.
---
## Assumption 4: No Perfect Multicolinearity
- Consider the following model
`$$wage_{it} = \beta_0 + \beta_1 experience_{it} + \beta_2 male_{i} + \beta_3 white_{i} + \alpha_i + u_{it}$$`
- `\(experience_{it}\)` measures how many years worker `\(i\)` has worked before at time `\(t\)`.
- Multicollinearity issue because of `\(male_{i}\)` and `\(white_{i}\)`.
- Intuitively, we cannot estimate the coefficient `\(\beta_2\)` and `\(\beta_3\)` because those **time-invariant variables are captured by the unit fixed effect `\(\alpha_i\)`**.
---
class: title-slide-section, center, middle
name: logistics
# Estimation
---
## Estimation with Fixed Effects
- Can estimate the model by adding dummy variables for each individual.
- **least square dummy variables (LSDV) estimator**.
- Computationally demanding with many cross-sectional units
- We often use the following **within transformation**.
---
## Estimation by within transformation
- Define the new variable `\(\tilde{Y}_{it}\)` as
`$$\tilde{Y}_{it} = Y_{it} - \bar{Y}_i$$`
where `\(\bar{Y}_i = \frac{1}{T} \sum_{t=1}^T Y_{it}\)`.
- Applying the within transformation, can eliminate the unit FE `\(\alpha_i\)`
`$$\tilde{Y}_{it} = \beta' \tilde{X}_{it} + \tilde{u}_{it}$$`
- Apply the OLS estimator to the above equation!.
---
## Importance of within variation in estimation
- The variation of the explanatory variable is key for precise estimation.
- Within transformation eliminates the time-invariant unobserved factor,
- a large source of endogeneity in many situations.
- But, within transformation also absorbs the variation of `\(X_{it}\)`.
- Remember that
`$$\tilde{X}_{it} = X_{it} - \bar{X}_i$$`
- The transformed variable `\(\tilde{X}_{it}\)` has the variation over time `\(t\)` within unit `\(i\)`.
- If `\(X_{it}\)` is fixed over time within unit `\(i\)`, `\(\tilde{X}_{it} = 0\)`, so that no variation.
---
## Various Fixed Effects
- You can also add **time fixed effects (FE)**
`$$y_{it} = \beta' x_{it} + \alpha_i + \gamma_t + u_{it}$$`
- The regression above controls for both **time-invariant individual heterogeneity** and **(unobserved) aggregate year shock**.
- Panel data is useful to capture various unobserved shock by including fixed effects.
---
## Cluster-Robust Standard Errors
- In OLS, we considered two types of error structures:
1. Homoskedasticity `\(Var(u_i) = \sigma^2\)`
2. Heteroskedasitcity `\(Var(u_i | x_i) = \sigma(x_i)\)`
- They assume the independence between observations, that is `\(Cov(u_{i},u_{i'} ) = 0\)`.
- In the panel data setting, we need to consider the **autocorrelation (自己相関)**.
- the correlation between `\(u_{it}\)` and `\(u_{it'}\)` across periods for each individual `\(i\)`.
- **Cluster-robust standard error (クラスターに頑健な標準誤差)** considers such autocorrelation.
- The cluster is unit `\(i\)`. The errors within cluster are allowed to be correlated.
</textarea>
<style data-target="print-only">@media screen {.remark-slide-container{display:block;}.remark-slide-scaler{box-shadow:none;}}</style>
<script src="https://remarkjs.com/downloads/remark-latest.min.js"></script>
<script>var slideshow = remark.create({
"countIncrementalSlides": false,
"highlightLines": true,
"highlightStyle": "github",
"ratio": "16:9"
});
if (window.HTMLWidgets) slideshow.on('afterShowSlide', function (slide) {
window.dispatchEvent(new Event('resize'));
});
(function(d) {
var s = d.createElement("style"), r = d.querySelector(".remark-slide-scaler");
if (!r) return;
s.type = "text/css"; s.innerHTML = "@page {size: " + r.style.width + " " + r.style.height +"; }";
d.head.appendChild(s);
})(document);
(function(d) {
var el = d.getElementsByClassName("remark-slides-area");
if (!el) return;
var slide, slides = slideshow.getSlides(), els = el[0].children;
for (var i = 1; i < slides.length; i++) {
slide = slides[i];
if (slide.properties.continued === "true" || slide.properties.count === "false") {
els[i - 1].className += ' has-continuation';
}
}
var s = d.createElement("style");
s.type = "text/css"; s.innerHTML = "@media print { .has-continuation { display: none; } }";
d.head.appendChild(s);
})(document);
// delete the temporary CSS (for displaying all slides initially) when the user
// starts to view slides
(function() {
var deleted = false;
slideshow.on('beforeShowSlide', function(slide) {
if (deleted) return;
var sheets = document.styleSheets, node;
for (var i = 0; i < sheets.length; i++) {
node = sheets[i].ownerNode;
if (node.dataset["target"] !== "print-only") continue;
node.parentNode.removeChild(node);
}
deleted = true;
});
})();
(function() {
"use strict"
// Replace <script> tags in slides area to make them executable
var scripts = document.querySelectorAll(
'.remark-slides-area .remark-slide-container script'
);
if (!scripts.length) return;
for (var i = 0; i < scripts.length; i++) {
var s = document.createElement('script');
var code = document.createTextNode(scripts[i].textContent);
s.appendChild(code);
var scriptAttrs = scripts[i].attributes;
for (var j = 0; j < scriptAttrs.length; j++) {
s.setAttribute(scriptAttrs[j].name, scriptAttrs[j].value);
}
scripts[i].parentElement.replaceChild(s, scripts[i]);
}
})();
(function() {
var links = document.getElementsByTagName('a');
for (var i = 0; i < links.length; i++) {
if (/^(https?:)?\/\//.test(links[i].getAttribute('href'))) {
links[i].target = '_blank';
}
}
})();
// adds .remark-code-has-line-highlighted class to <pre> parent elements
// of code chunks containing highlighted lines with class .remark-code-line-highlighted
(function(d) {
const hlines = d.querySelectorAll('.remark-code-line-highlighted');
const preParents = [];
const findPreParent = function(line, p = 0) {
if (p > 1) return null; // traverse up no further than grandparent
const el = line.parentElement;
return el.tagName === "PRE" ? el : findPreParent(el, ++p);
};
for (let line of hlines) {
let pre = findPreParent(line);
if (pre && !preParents.includes(pre)) preParents.push(pre);
}
preParents.forEach(p => p.classList.add("remark-code-has-line-highlighted"));
})(document);</script>
<script>
slideshow._releaseMath = function(el) {
var i, text, code, codes = el.getElementsByTagName('code');
for (i = 0; i < codes.length;) {
code = codes[i];
if (code.parentNode.tagName !== 'PRE' && code.childElementCount === 0) {
text = code.textContent;
if (/^\\\((.|\s)+\\\)$/.test(text) || /^\\\[(.|\s)+\\\]$/.test(text) ||
/^\$\$(.|\s)+\$\$$/.test(text) ||
/^\\begin\{([^}]+)\}(.|\s)+\\end\{[^}]+\}$/.test(text)) {
code.outerHTML = code.innerHTML; // remove <code></code>
continue;
}
}
i++;
}
};
slideshow._releaseMath(document);
</script>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
var script = document.createElement('script');
script.type = 'text/javascript';
script.src = 'https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-MML-AM_CHTML';
if (location.protocol !== 'file:' && /^https?:/.test(script.src))
script.src = script.src.replace(/^https?:/, '');
document.getElementsByTagName('head')[0].appendChild(script);
})();
</script>
</body>
</html>