10

A more apples to apples comparison would be between (i) Fama-Macbeth procedure and (2) clustering standard-errors by date. Adding fixed-effects is somewhat different. Problem: cross-sectional correlation causes naively computed standard errors to be understated Let $r_{it}$ denote the return of firm $i$ in month $t$. An important statistical issue is that ...


6

It probably does not have a mean or a variance. Ratio variables often don't. As these are accounting ratios, there are several candidate distributions and their ratios wouldn't have a first moment. There are a couple of things to remember. If none of the variables in a ratio can be negative, then you likely have a truncated distribution. In practice, this ...


2

Excluding an autoregressive term from a regression (a regular univariate time series regression or as in your case a static panel data regression) is not an omitted variables problem. For the univariate case most standard textbooks in statistics cover that autocorrelation in the residuals still leaves the OLS estimate unbiased and consistent (but inference ...


2

It means that about 25 percent of your within firm variation (i.e. the variation of a firm over time) is explained by your time-series dummies i.obs which pickup the period by period cross-sectional mean. To phrase it another way, about 25 percent of a firm's variation in $y_{it}$ over time is explained by variation in the aggregate $\bar{y}_t$.


2

I think a useful paper would be Rapach et al. (2013), International Stock Return Predictability what is the role of the united states? They detail cross country differences in a lot of detail and tell you what you need to consider. In terms of time differences etc. It is also an easy to paper to digest and there is a summary on CFA digest.


1

Generally, you need a industry classification scheme to identify each company's industry. For U.S. and Canadian firms, SIC are a reasonable choice, since they are broadly available. For example, you can find SIC codes for U.S. and Canadian firms here: https://siccode.com/ Alternatively, researchers could use MSCI industry classifications or comparable ...


1

As part of your analysis, it is always a good idea to do something simple before pulling out the big guns. So, OLS by country with perhaps a handful of controls would be a good benchmark, if only to tell later if your complicated ideas don't amount to squashing a fly with a sledge hammer. As for the panel idea, you have to think about how you'd be pooling ...


1

What you try to do is not clear. If you want to reimplement FF factors on a custom universe, you can do it (but paying attention to timezones to not include future information in your data). But be careful: your result will only show you if there is a cross-sectional Factorial effect inside your firms. May be you want in reality see if your firms have a ...


1

I am actually unsure of your question. Both the fixed effects and random effect models will depend on the assumptions that you make. To keep it simple, in the Fixed effects model: all the individual differences are captured by differences in the intercept parameter. In the Random effects model: all individual differences are captured by the intercept ...


1

If there is autocorrelation than you need to add the lagged dependent variable. By not including it, your regression is suffering from the omitted variable bias. You say that by doing this you will be "modelling liquidity where liquidity of the previous day is the most important factor" but since your regression "demands" adding the LDV (due to the AC) then ...


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