- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello everyone,
I am currently conducting an empirical study as a part of my master thesis on the topic Cash Flow Forecasts. So I basically try different models and compare which model has the better explanatory power. I am performing OLS regressions cross- sectionally. However, since most of the models are univariate I want to (i) check if the parameter estimates of the different ratios are statistically different from each other and (ii) perform pair-wise tests to check which parameter estimate is more persistent than the others.
Any suggestions how I can perform this tasks?
Thanks in advance.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
However, since most of the models are univariate I want to (i) check if the parameter estimates of the different ratios are statistically different from each other and (ii) perform pair-wise tests to check which parameter estimate is more persistent than the others.
"However, since most of the models are univariate..." I think I understand this part (except for the word "most", that isn't clear)
"I want to (i) check if the parameter estimates of the different ratios are statistically different from each other ..." This part is not clear to me. Do you mean you have a fitted model Yhat=b01+b1X1 and another model Yhat=b02+b2X2 and you want to compare b1 to b2?
"... (ii) perform pair-wise tests to check which parameter estimate is more persistent than the others." I have no guess as to what this means.
Earlier, you said "I basically try different models and compare which model has the better explanatory power", where does this fit in with the above? Testing parameter estimates, as you discuss above, is not the same thing as finding which model has the better explanatory power.
Paige Miller
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Sorry about that. This is, of course, a different check I want to perform to be consistent with previous literature on cash flow forecasting. So I post here two snapshots (from Barth et al. (2001) for (i) and Cheng and Hollie (2008) for (ii)) which I believe describe hat I want to do.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
@NikolayDR wrote:
Sorry about that. This is, of course, a different check I want to perform to be consistent with previous literature on cash flow forecasting. So I post here two snapshots (from Barth et al. (2001) for (i) and Cheng and Hollie (2008) for (ii)) which I believe describe hat I want to do.
Do either of those papers go into sufficient detail for the comparisons you want?
I am afraid that a small text extract like this and no actual link to the papers isn't terribly helpful.
I am not an economist and don't even play one on TV. So without a concrete example of what is done I am clueless for specific jargon laden text like that.
I do know several "pair-wise" comparison techniques. Some of them are pretty flaky but they are comparisons. Doesn't mean I would suggest them as statistically valid or useful without supporting details.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
@NikolayDR wrote:
Sorry about that. This is, of course, a different check I want to perform to be consistent with previous literature on cash flow forecasting. So I post here two snapshots (from Barth et al. (2001) for (i) and Cheng and Hollie (2008) for (ii)) which I believe describe hat I want to do.
This just re-phrases the issues I don't understand in different words, but it provides no additional information and hence no additional understanding. Can you explain these items in detail?? Can you give examples of each?
Also, I asked a very specific question about part 1, which you haven't answered.
Paige Miller