Sign up  |  Log in

Study Session 3: Quantitative Methods for Valuation

Durbin Watson for Serial Correlation, WHY?

I rushed through the quant material when I took the exam last year. I am no quant expert, but this question has been bugging me for some time. 

Why does the CFA decide to go with Durbin Watson to test for serial correlation? I thought DW only tests for first order serial correlation and the inconclusive area of the test makes it a less effective test. Is there a reason behind this? 

Degrees of freedom

Degrees of freedom, according to my understanding should be n-1. But as I am observing in most of the questions (Schweser question bank), they have taken them as n-2. Any comment or clarification, would help me.

Correlation and Regression

Could anyone explain/provide a mathematical proof for the following statement?

‘If the units of the independent variable are tons instead of pounds, the estimated slope coefficient will be 2000 times larger’

MM and/or IFT?

Hi all,

I just started L2 and decided to start with quant. I am using MM’s videos first and then reading the CFAI text. So far this approach has worked with the more qualitative readings (fin tech). But when I took this approach with correlation and regression, I have trouble following MM’s videos and the CFAI reading is dense and confusing as heck.

AR Models

Hi guys! While I was doing some questions of the Time-Series Analysis part, I came up with a question which I can’t fully justify de answer. The question says:

Which of the following AR models is most appropiate for a time series with annual seasonality using quarterly observations?

a) b1xt-1 + b2xt-12​ + et​​​​​​

b) b0 + b1xt-1 + b2xt-4​ + et​​​​​

Covariance stationary time series


I understand that if both the intercept and slope coefficient in a time series do not differ significantly from 0 then it is a covariance stationary time series but how can we make valid inferences if both are not statistically significant? Does the fact that the time series is covariance stationary take precedence over the fact that the regression coefficients are statistically insignificant? Would we still be able to make proper statistical inferences?



Is this understanding correct?

Regression SS is the sum of squares that is explained by your model. Residual SS is the sum of squares that is not explained by your model. 

Random Walk

@Harrogath - I got this from one of the past posts. I did not understand - how past value is the best predictor of the value today when you cannot forecast a Random walk?


Durbin-Watson Rejection Points


When we conduct the DW test, we fail to reject the null if DW<dl. This is a case where positive serial correlation is present. When negative serial correlation is present, we reject if DW>4-dl. Why 4-dl?

thanks your help!