<t>A t-test the significance of a variable individually.,The F-test test the significance of a set of variables jointly. So if our model was y = x1 + x2 + error terms. We would have two T-test results and one F test result. The F-test of all the variables (i.e. X1 and X2) is equivalent to testing th...

<t>As for macro modelsStructural based models tend to be a whole system of equations based on various economic relationships. These systems would tie in the monetary sector, real sector, etc, etc in a whole system of equations. Just think of it as multiple regressions run simultaneously. Some calibr...

<t>Econometric methods tend to be either Time series models (under ARIMA, Cointegaration method) or Cross Sectional (Latent variable, Logit...regression method).You are either studying a variable across time or you are studying a person/country/? behaviour across several variables.Panel methods allo...

<t>How are the Stalla Notes compared with Scheswers?Not sure, since i never used Stella -suppose to be a bit tougher on the questions. i think.Schweser is fine as I used it at Level 1 and 2., Stopped there since I realised the CFA was quite useless in my line of work.The knowledge of course always h...

<t>materials should be enough, but for areas where you may be weak at e.g. accounting get the textbook as added material.Since accounting is such a big component - good to review that as an extra. I assume stats, derivatives is a a walk in the park for you since they are so basic. BTW the questions ...

<t>If i am not mistaken a log variable also has a different interpretation. The variable is no longer interpreted at the levels but as a rate of change(%). So for a semi log function of Y = ln(X1) + c. For a 1% change in X1 leads to a x unit change in Y. We usually logs when we wish to interpret the...

<t>It depends on the objective. If the objective is to get overall company revenue. Then adding all the values at once would be better - Occam's Razor - simpler is better.However if your objective is to understand the individual product revenue dynamics - especially if you believe that the true rela...

<t>I;ve done the same research before, and your question made me go back and rethink what my results were.My results showed that the intercept played a very strong role. both in terms of the size of the coefficient and the T-stats significance level.If I recall upon deeper investigation, we found th...

<t>I think RIskCalc uses a non-linear logit - which is really many univariate-like logit regressions and then combined. the probit is a normal dist and logit is a logistic dist. the logistic dist has slighltly fatter tails than the normal distribution.As for the output - there is no practical differ...

<t>You cannot do multinomial logir under Eviews. Eviews is predominantly a time series econometrics package. Less suitable for multinomial and nested logit models. LIMDEP is one of the predominant packages for multinomial logit models. You can try SAS or SPSS as well. (but you have to check)I would ...

<t>>What about historical testing/backtesting? Well just cause the model was successful in predicting the past does not mean it will perform as well predicting the future. Certain events may not have occured which would have caused the model to fall apart. Models backtested before the asian crisis a...

<t>I need to develop specs for a risk data mart and review some vendor datamarts.My concern is their adequacy of their data models. Many IT companies may not necessarily have thought throught the business and operational issues. Since I'm not in the best position to know about data models, does anyo...

<t>HSBC in Malaysia are getting to be pretty active in it. I hear Citibank was active in it, but decided to delay post 9/11, but HSBC quick surge prompted a lot of foreign banks including Citi to jump start and get moving again. StandChart has started to move in as well. In general, M'sian market is...

<t>70% does give room for healthy skepticism. But I would think keeping and open mind is important. So maybe they exagerrated and the predictive accuracy within the bands was 60% instead. That's still makes it a useful too. I wouldn't say that all such technologies overfit.Don;t throw the baby out w...