Business Forecasting Topic 8 Flashcards

1
Q

Box Jenkins

A

ST forecasting
- exploit correlation between observations at different points in time

  1. identify tentative model
  2. use series of diagnostic tests to assess model adequacy
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

stationarity (time series)

A

structure doenst change over time

each distribution = same mean and variance irrespective of time period

mean and variance = constant over whole time span of series

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

autocorrelation

A

correlation of a variable with itself at different times
-graphed above confidence limit = significant -> confidence limit requires std dev and mean

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

partial autocorrelation

A

how much of a correlation is there between sales k-periods apart, when the effects of the intervening weeks are removed

pattern of PACFs = help us identify forecasting model appropriate for given time series

above confidence limit = significant
influence of intermediary things are removed

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

significance tests

A

applied to ACF and PACF based on null hypothesis that data is completely random

approx standard error = 1/ root of n (n = number of data points)

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

calculating the ACF or PACF coefficients

A

significant values lie outside

these values are plotted as confidence limits on the diagrams

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

autoregressive models

A

first order autoregressive model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

ACF AND PACF for autoregressive models

A

ACF - declines rapidly to zero (exponentially) or oscillates between positive and negative values and dying

PACF- cuts off after lag 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

moving average models

A
  • show relationship between variable to be forecast shocks in earlier periods
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

ACF AND PACF for moving average models

A

exact reverse of first order autoregressive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

mixed models

A
  • involve both autoregressive and moving average terms
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

ACF AND PACF for mixed models

A

both decline rapidly to zero

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

ARIMA

A
  • classify different models that are available
    ARIMA (p, d, q)

ARIMA (1,0,0) = first order autoregressive
ARIMA (0,0,2) = second order moving average
ARIMA (1,0,1) = mixed model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

degree of differencing

A

stationary series require no differencing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

stages in box Jenkins approach to forecasting

A
  1. identify appropriate model
  2. estimate model parameters
  3. use diagnostic tests to make sure these parameters are appropriate
  4. use model to make forecasts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

series of 4 diagnostic checks to test adequacy

A
  1. are residuals of model white noise
  2. residuals approximately normally distributed
  3. coefficient in model significantly different from zero
  4. overfitting
17
Q

residuals white noise

A

residual = actual sales = model prediction

not white noise = still some pattern in series that model isn’t picking up (pattern useful for forecast)

below confidence limit = white noise

18
Q

residuals approx normally distributed

A

assumption necessary for t test

19
Q

coefficient in model significantly different from zero

A

if not = suggest constant is 0 and or previous sales provide no useful info for this week

p value below 0.05 - significantly different from zero

20
Q

overfitting

A

fitting series of more complex models to data and seeing if coefficients for extra terms are sig different from zero

p value > 0.05 = coefficient is not sig diff from zero - indicates original model probs adequate for time series

21
Q

making forecasts

A
  • forecasts can be expressed in form of confidence intervals
  • forecasting from ARIMA (0,0,1) = more problems need to know error (shock) in forecasting final figures