Recent content by kiton

Thank you @hlsmith. Confidence intervals are exactly what I am after. I'd appreciate the update on the name.

Thank you for response @Miner. I will look into discrete event simulation.

Hello dear forum members, I am seeking your feedback on the question that I have limited knowledge about (i.e., stochastic modeling). Say, I am working with a variant of standard epidemiological compartment model -- SEIR (Susceptible --> Exposed --> Infected --> Recovered): In my analysis I...

Following the same Google example, people started rating restaurants many years ago, from 2008 I believe. Some restaurants mights have as low as 1 review, while others 2000+. For each of the reviews only year is available (no specific date); therefore, individual posts are captured at the yearly...

@hlsmith Okay, let me try to explain it this way. The closest comparison to the situation I am trying to study would be this. Google has a service called Google Reviews. People often use these, for instance, when searching for a restaurant. Here is an example of such reviews. Look at the...

@Miner You are surely right on this. I plan to address endogeneity caused by self-selection once I have the main model justified.

@hlsmith I appreciate your feedback. Let me clarify. The platform provides user ratings of the companies. Any registered platform user can rate any company that has a registered account. In order to improve the quality of ratings, the platform introduced a new user-level feature that allowed...

I hear you. Then MIDAS seems to me the only feasible option. Admittedly, however, I have very limited knowledge about it and there is only one paper on this in my library.

Mediation analysis is a completely different thing, which by itself is irrelevant to your estimator. You can look here to get a sense of it.

Your data structure is somewhat unclear to me... What is the measurement of your outcome -- is it seconds (or minutes) it takes an individual to answer the whole questionnaire or one single question?

Anything, but Excel :) Sorry.

If your data are recorded for a given fiscal period (i.e., one value for a given period), I don't see any feasible ways to expand it, it is what it is. Or am I missing something here?

I suggest you take a look at the following material to better understand the difference between fixed and random effects. In brief, you should (a) have some theoretical reasoning describing relationships between your time-variant and invariant predictors, then (b) run both FE and RE models and...

Can you aggregate your daily data at monthly level and check consistency of your estimates with added IVs?

What kind of model are you using? Is it a fixed effect model or something else? Also, how are you treating your year variable as a numerical or categorical?