Time Series Analysis

Question 1

Regression of a variable observed at different time periods.

Answer 1

time series analysis

Question 2

limitation of time series analysis

Answer 2

1. regression assumptions are unlikely to be met - e.g. residual errors are correlated
2. the mean and variance may change over time

Question 3

simplest time series

Answer 3

trend

Question 4

simplest trend model

Answer 4

linear trend - dependent variable Y changes constantly with time (independent variable)

Question 5

Why use a log-linear model?

Answer 5

If the residuals persist

Question 6

what does a log linear model show?

Answer 6

exponential growth which has had a log applied to it, thus showcasing growth linearly

Question 7

test for serially correlated errors

Answer 7

durbin watson test

Question 8

time series model that regresses its own lags (past values)

Answer 8

autoregressive series

Question 9

What is a covariant stationary series?

Answer 9

Time series where mean and variance remain the same throughtout

Question 10

3 requirements of a covariant stationary time series

Answer 10

1. expected value of the time series is constant and finite
2. variance constant and finite
3. covariance constant and finite

Question 11

What is the standard error of autocorrelation?

Answer 11

1/sq rt(T) where t is the number of periods in data

Question 12

series roughly stays within an average

Answer 12

mean reversion

Question 13

What is the chain rule?

Answer 13

Use mean reversion to estimate the next period or 2 of values in a time series

Question 14

whats the limitation of the chain rule?

Answer 14

It increases uncertainty because future estimates are used to calculate other values

Question 15

two types of forecast errors?

Answer 15

in and out of sample forecast errors

Question 16

why is out of sample forecast errors the preferred method?

Answer 16

allows precision of the forecasted values to be assessed - compares predicted values with actual values

Question 17

guidelines for instability of regression coefficients

Answer 17

1. dont mix data from 2 periods with different types of regulation

Question 18

What is a random walk?

Answer 18

A time series with a predictable component and a random error component.

Question 19

what time series smooths noise?

Answer 19

moving average

Question 20

What is an autoregressive time series?

Answer 20

One that predicts future values based on past data.

Question 21

What is a moving average time series?

Answer 21

Random noise fluctuations are smoothed over time to reveal a trend.

Question 22

What is an ARMA model?

Answer 22

auto regressive moving average

Question 23

What is the minimum data for ARMA?

Answer 23

80 observations

Question 24

What is the ARCH model?

Answer 24

auto regressive conditional heteroskedasticity

Question 25

5 possible scenarios with multiple time series

Answer 25

1. neither time series has a unit root (use linear regression)
2. reject null of unit root for independent variable (should not use linear regression)
3. reject null of unit root for dependent variable (should not use linear regression)
4. both series have unit roots but not cointegrated (should not use linear regression)
5. both series have unit roots but are cointegrated (should not use linear regression)