QUESTION 1

kurtosis equal to 1 implies heavy/fat tails

Yes

No

QUESTION 2

Fat tailed distributions have infinite/indefinite variance

Yes

No

QUESTION 3

Pick the fat tailed distributions in the list

Normal (Gaussian)

Poisson

Student t

Cauchy

Pareto

GEV

Log normal

Uniform

Binomial

QUESTION 4

Pick the distributions in the list that are best suited to model the maximum value of the random sample from a distribution

Normal (Gaussian)

Poisson

Student t

Cauchy

Pareto

GEV

Log normal

Uniform

Binomial

QUESTION 5

Pick the stable distributions in the list

Normal (Gaussian)

Poisson

Student t

Cauchy

Pareto

GEV

Log normal

Uniform

Binomial

QUESTION 6

If A and B are random variables from the same stable distribution then

the new variable C=A-B is from the same distribution but possibly with a different location and scale

the new variable C=A+B is from the same distribution

the new variable C=A-B may not be from a stable distribution

the new variable C=A+B is from normal distribution but possibly with a different location and scale

QUESTION 7

Stable distributions are a generally good fit for stock returns distributions

Yes

No

QUESTION 8

Volatility clustering models even with Normal disturbances can produce heavy tails

Yes

No

QUESTION 9

In financial time series analysis unlike in Economics the tail dependence is not an issue

Yes

No

QUESTION 10

Do not use Gaussian copula to model financial time series because it always produces zero tail dependence

Agree

Disagree

QUESTION 11

Bandwidth choice is more important than the Kernel shape choice in application of KDE

Yes

No

QUESTION 12

Difficulty of density is not an important factor in KDE, all distributions have approximately the same difficulty

Yes

No

QUESTION 13

Epanechnikov kernel is more efficient than Normal but it doesn't mean we should always use it

Yes

No

QUESTION 14

Local KDE method uses constant bandwidth

Yes

No

QUESTION 15

Johnson SU distribution should be used for empirical data exhibiting low kurtosis

Yes

No

QUESTION 16

pick The distributions that are best choices for the body of the piecewise density

pareto

GEV

Kernel Density

ECDF

QUESTION 17

pick The distributions that are best choices for the tails of the piecewise density

pareto

GEV

Kernel Density

Normal

QUESTION 18

Kernel regression problem is formulated as a weighted least squares problem

Yes

No