There is a way of colculation a future standart diviation of a portfolio if you know the daily standart diviation and the mean.

http://www.stat.berkeley.edu/~aldous.../Ugrad/ZY3.pdf

page 2 last formula.

the problem i have is how to calculate the standart diviation of the same portfolio with deposits.

lets say for example that i have a portfolio with a daily standart diviation of 1% and a daily mean return on 0.05%. i am starting with 1000$ and i add 1$ every day for 500 days. what will be the var or the standart diviation of the sum after 500 days.

i dont need the answer but the formula, if posible plese refer me to a paper on that matter. i need it for some calculation i am trying to do at work. meanwhile i am using a Monte-Carlo simulation and it is taking too muth time and it is not acurate....

thank you, please help. ]]>

In my model I include three sets of fixed effects. Specifically they cover origin countries (25), and industries (22) and Years (7), in total 52 dummy variables. I have tried calculating VIF values with- and without these dummies. Sample size is around 6000. Its a cross-sectional analysis (not panel data).

When calculating VIF with the dummies, tolerances are acceptable, all with VIF < 4. However when adding the dummies, tolerance values of ALL predictors drop to very low values (even the lowest VIF is > 10), indicating the predictor variables that I want to use for inference is basically useless (at least for explanatory analysis of these, as I understand it). The non dummy variables (besides a single pair) is not correlated:

Using SAS REG and GLM (automatic fixed effects dummies not possible in REG) respectively:

I have tried adding and removing variables to see if the problem was due to correlations between the predictor variables (not the dummies), but I can conclude that the VIF inflation happens due to the dummies.

Theoretically it makes sense to add the dummies to account for non measured country, industry and time dependent effects. Also the variables included makes theoretical sense although they may be correlated. E.g. the number of foreign subsidiaries of a firm is often correlated to its age and size, but still theoretically the effect of these variables is different in regards to the dependent variable. Is it possible to create model in which I can validly explain relationships between independent and the dependent variable if VIF values are high like this? If not what can I do besides omitting important predictors? ]]>

I work in a university and am trying to analyse a dataset that contains data for a number of years on whether students perform sufficiently well in their first year exams to progress to the next year of their programme in the next academic year (yes/no)

The factors that I want to analyse are all categorical, as follows:

Student Domicile – Home or Overseas

Student Ethnicity – Blank/Minority/Ethnic (BME) or Not BME

Student application route – UCAS or Clearing

Student in Receipt of Bursary – Yes or No

If I just wanted to look at the relationship between one of these variable and whether the student progresses or not then I could use chi-square, but the tables here are more than 2x2 I understand that log linear analysis is the technique I should use.

What I am looking for is a good worked example of an approach to explore the relationships between these five variables = Domicile+Ethincity+Route+Bursary+Progression.

Can anyone suggest one please? Preferably one that is based on SPSS

Ultimately I want to build a probability calculator to calculate the likelihood of a student progressing depending on these other variables.

Thanks very much,

StatsLearner ]]>