Ch 16 - Modelling high frequency data : Irregularly spaced data Flashcards

1
Q

What is the objective of modelling durations?

A

To forecast the arrival time of the next observation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is a trade duration?

A

The time between 2 trades, usually measured in seconds

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Periods of intense trading are generally periods of greater market _______ than periods of sparse trading

A

Liquidity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Autocorrelation in raw durations is _____ than in deseasonalised durations due to the diurnal patterns in the durations

A

Higher

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What 3 parts are durations decomposed into?

A

A deterministic diurnal component
An innovation term
A time varying conditional mean term

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How is the ACD model estimated?

A

Via MLE using an exponential distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is one way to formally evaluate the accuracy of the ACD model?

A

Mincer-Zarnowitz test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly