# Tag Info

7

Consider the following regression specification where, $t$ is time, $c$ is the firm, $y$ is an outcome and $x$ is a variable of interest. $$y_{c,t} = \alpha + \beta x_{c,t} + \varepsilon_{c,t}$$ There are three types of omitted variables: Variables that vary with time but are the same across firms. Examples might be weather conditions, inflation rate, ...

6

Using corruption is part of it but a bit restrictive way to measure government "quality". You may use aggregate indicators as the one developed by the Worldwide Governance Indicators (WGI) project from the World Bank. They reports aggregate and individual governance indicators for over 200 countries and territories over the period 1996–, for six dimensions ...

5

Use -areg- in Stata, and the standard errors will come out as in the textbook. Specifically, the command areg lpassen lfare ldist ldistsq y98 y99 y00, absorb(id) vce(robust) will produce the desired result. -xtreg- with fixed effects and the -vce(robust)- option will automatically give standard errors clustered at the id level, whereas -areg- with -vce(...

4

I'm still not sure if I'm doing something wrong. However, it is useful to note that I get the same results in R. library(foreign) library(plm) library(lmtest) df <- read.dta("airfare.dta") fe.out <- plm(lpassen ~ lfare + ldist + ldistsq + y98 + y99 + y00, data=df, index = c("id", "year"), method = "within", effect = "individual") ...

4

This question is related to a post I addressed on CrossValidated. The "generalized" difference-in-differences (DiD) estimator is amenable to settings with multiple groups and multiple exposure periods. Take the following specification: $$y_{it} = \gamma_{i} + \lambda_{t} + \delta T_{it} + \epsilon_{it},$$ where $\gamma_{i}$ and $\lambda_{t}$ ...

4

I do not think the premise is correct. Following Brüderland and Volker in Best & Wolf The SAGE Handbook of Regression Analysis and Causal Inference [square brackets have my remarks]: Both estimators require strict exogeneity [Fixed Effects (FE) and First Differences (FD)]. However, while FE builds on the assumption of no serial correlation prior to ...

3

Here the solution would depend on what you want to accomplish. Note the problem is not just that the series is unbalanced, for an ordinary unbalanced panel data-set where firms have different number of $T$ observation the command would still work. Here no adjustments are necessary, you can easily try it yourself: install.packages("plm") library("plm") data(...

3

Having read up on your question it seems the fixed effect is fixed. If this is indeed the case it will have zero variance and hence zero covariance with any variable.

3

I cannot precisely answer your questions because I do not know which exactly regressions you want to perform as @jmbejara says and which papers are you referring to that use Fama-MacBeth regression. Are they on financial or other literatures? I haven't seen Fama-MacBeth on other literatures (I do not follow any other literature to be exact), so please post ...

3

The technique described in the question is almost correct. Consider a panel data set consisting of three cross-sections ($a$, $b$, and $c$) and three time-periods ($1$, $2$, and $3$). Let y denote the column vector with the observations of the dependent variable, x the column vector with observations of the first explanatory variable, and z the column vector ...

3

To understand the issue let's review what is the so call robust variance-covariance matrix estimates (VCE) and the implied "robust" standard errors. The robustness is meant to allow for violations of homoscedasticity in the cross-sectional dimension or heteroscedasticity. There are various heteroscedastic robust VCE which are known as the Sandwich estimators ...

3

FE logit requires the idiosyncratic errors to be IID across $i$ and $t$, quite a strong assumption. Also the regressors should be strictly exogenous, but it's the same for linear FE models. In your application, the fact that FE logit wouldn't converge will make a good argument against FE logit, and will satisfy some referees but not all. An important ...

3

The constant term in your final FE model has no specific meaning without further restrictions. For Stata, it is only chosen such that the (sample) mean of the estimated individual effects add up to 0. So your testing $\alpha>0$ is in fact based on $$\frac{1}{n} \sum_{i=1}^n \hat\alpha_i,\quad \hat\alpha_i = \frac{1}{T_i} \sum_{t=1}^{T_i} (x_{it} - y_{it}),... 3 As rightly pointed out by @1muflon1@ "Panel data is nowadays quite a big field - usually you will have separate chapters for panel IV, panel logit/probit, panel time series etc". But if you are "looking for a briefer introduction/overview", I would recommend: Econometrics Training: Module Three Panel Data (with William Greene and John ... 3 Here is an example where just from an economic perspective fixed effects are better than random effects. Suppose you have panel data and you want to regress earnings y on some observable characteristics X of an individual like education, tenure, experience, age, birthplace, etc. The regression you would estimate is$$y_{it} = \alpha + X'_{it} \beta + \...

3

Your model has $\beta_3 * t$, which is a linear time trend, not time dummies. If that's correct, you are controlling for only a linear trend. Because oil prices do not have a perfect linear trend, you can include them. But I am not sure you really want the linear trend specification instead of time dummies (say, $\beta_{3t}$). For a model with common time ...

3

I've seen them summed somewhere but I cannot exactly remember where. Ultimately I don't think that it makes much difference. The quarterly sum is just the average multiplied by three. Since local projections are just a bunch of OLS - one for each horizon, this is how you can think about the issue: If $y_{t+h} = \alpha^h + \beta_h news\_shock_t + \sum_{j=1}^{... 3 [W]hy do they need to write down "adopted a leniency law at some later point of time"? Because in Korea case, the word "our sample period" means "1995-2002" already. Assuming Korea is the early-adopter country, then all countries theretofore untreated before 1997 may serve as a counterfactual. This includes the countries never ... 3 tldr If$\mathbb{E}[\varepsilon_{i,t}|X_{i,t-1}, Y_{i,t-1}] = 0$then the coefficient$\beta_2$is equal to: $$\frac{\partial \mathbb{E}[\ln Y_{i,t}|X_{i,t-1}= x_{i,t-1}, Y_{i,t-1} = y_{i,t-1}]}{\partial \ln x_{i,t-1}}$$ If$\varepsilon_{i,t}$is independent of$X_{i,t-1}$and$Y_{i,t-1}$(which is a stronger condition) then it is also equal to: $$\frac{\... 3 The number of observation changes likely because there are missing observations for some controls. Unless you suspect that statistics might systematically not collect data for some firms (e.g. maybe data on some control was recorded only for big firms) this is not concerning. You could consider investigating if there are any systematic reasons for missing ... 2 By stnadard OLS regression results, in the simple regression$$y_t = \alpha + \beta z_t + v_t, \;\;\;t=1,...,T$$we have that$$\hat \beta = \frac {\sum_{i=1}^T (z_t - \bar z)(y_t-\bar y)}{\sum_{i=1}^T (z_t - \bar z)^2}$$and$$\hat \alpha = \bar y - \hat \beta \bar z$$So the residuals are$$\hat v_t = y_t -(\hat \alpha +\hat \beta z_t)$$Then, ... 2 You have done two different things. Your fixed-effects model captures the within-group over-time functional relationship between$debt_{it}$and$y_{it}$(that is, how much average difference in$y_{it}$is there between two periods with a 1-unit difference in$debt_{it}$within a country). In your data, there is limited within-group variability in$debt_{...

2

The data sample is so small that formal testing for stationarity would be essentially worthless. Inspect visually your individual series for any obvious trend. This would be the case where even with a short sample non-stationarity would be a problem.

2

You can't identify the effect of oil price when Year FE are applied, since the world oil price is perfectly correlated with year Fixed Effects. You can't identify the democracy indicator if you country does not change its value in your observed period. For the other variables, it should be possible to obtain estimates. You should apply a fairly large ...

2

You can regress residual squares (from RE or FE depending on your estimation) on $X_{it} \hat\beta$ and its square using the clustered standard errors (the vce(cl id) option), and read the F statistic and the associated p value. This is basically the same as Het test for cross sectional models (White's simplified test). xtreg y x1 x2, re predict uhat, ue ...

2

It would be helpful to provide a reproductible example. In the paper Panel Data Econometrics in R: The plm Package, the authors explicitly mention that economic panel datasets often happen to be unbalanced, which case needs some adaptation to the methods. Hopefully, they provide a solution and the result of their work is bundled in the plm add-on package. ...

2

Just to make sure I understand: You have a daily panel (with missing values), probably weekday only, running from 2013 to mid-2017 with $n=3$ cross-sectional units. You believe that after 2015, there is a stronger incentive to do something different at quarter end, so the average basis should be different on month end days during that period. The typical ...

2

Yes it does. According to the Verbeeks guide to modern econometrics (pp418) standard panel fixed effects binary model assumes that error has “a symmetric distribution with distribution function $F(.)$, i.i.d. across individuals and time and independent of all $x_{is}$” [with the model being $y_{it}^*=x_{it}’ \beta+ \alpha_i+u_{it}$]. Just a two pages later ...

2

Note that this section in the book starts with the counterfactual that each agent has an outcome $y_0$ without treatment and an outcome $y_1$ with treatment. You observe only the $y_0$ of the untreated agents and only the $y_1$ of the treated agents, but both exist for each agent. If treatment is randomized across agents then $w$ is indeed independent of $(... 2 In panel regressions you have multiple dimensions and that is why also you have 3 different$R^2$. The within$R^2$tells you how much variation within your panel variables is on average explained by your model. The between$R^2$, tells you how much variation between your panel variables is explained by the model, and overall$R^2\$ gives you the combination ...

Only top voted, non community-wiki answers of a minimum length are eligible