Quant Shit Flashcards Preview

CFA 2 > Quant Shit > Flashcards

Flashcards in Quant Shit Deck (24):
1

Conditional heteroskedasticity is

residual variance related to level of X's

2

Serial correlation is

correlated residuals

3

Multicollinearity is

two or more X's are correlated

4

Effect of conditional heteroskedasticity

Type I errors
high t stat, caused by low std errors

5

Effect of serial correlation

Type I errors
positive correlation

6

Effect of multicollinearity

type II errors

7

Detection of conditional heteroskedasticity

Breusch-Pagan Test
Chi-Square Test

8

Detection of serial correlation

Durbin-Watson test

9

Detection of multicollinearity

Conflicting t and F stats
Correlations among ind variables if k=2

10

Correcting conditional skedasticity

white-correct std errors

11

Correction serial correlation

Hansen method

12

Correcting multicollinearity

Drop a correlated variable

13

Functional Form Misspecifications

-important variables omitted
-variables not transformed properly
-data pooled improperly

14

Time-Series Misspecification

-X is lagged Y with serial correlation present
-Forecasting the past
-Measurement error

15

Probit model

estimates probability of default given values of X based on normal dist

16

Logit Models

estimates probability of default given values of X based on logistic dist (computationally easier than normal dist).

Logistic dist NOT logarthimic

17

Discriminant models

produces a score or rank used to classify into categories
ex- bankrupt, not bankrupt

18

Economic Significance

not significant just because of statistical significance
-commissions, taxes, risk, etc.

19

If a time series is mean reverting

the value of the dependent variable tends to fall when above its mean; and rise when below its mean

20

Mean Reverting Level Formula

b0/ (1 - b1)

21

Forecasting Accuracy of ARCH measured by

root of mean squared error.
Use model with lowest RMSE based on out-of-sample forecasting

22

Without a mean reverting level, the time series is

non-stationary

23

Dickey-Fuller Tests for

unit root

24

Dickey Fuller Test method

subtract x(t-1) from both sides; first differencing
where g1 = (b1 - 1)

If there is a unit root in AR(1) model , g1 will be 0.