Week 6 Lecture 3 Forecasting Flashcards Preview

MSCI224 > Week 6 Lecture 3 Forecasting > Flashcards

Flashcards in Week 6 Lecture 3 Forecasting Deck (22)
Loading flashcards...
1
Q

Under what circumstances can linear regression be used?

A

When the data is linear and constant

2
Q

As opposed to the weight given to Moving Averages, the weight given in Exponential smoothing…

A

Changes and is determined by Alpha

3
Q

Under what circumstances should ‘Single Exponential Smoothing’ be used?

A

When the data is non-trended and non-seasonal

4
Q

The alpha used in Exponential Smooting is…

A

Not calculated. It is an estimate set beforehand.

5
Q

Between what values should Alpha lie?

A

Between 0 and 1

6
Q

What is the formula used in Single Exponential Smoothing for the calculation of St?

A

Alpha x Yt + (1-Alpha) x St-1

7
Q

Under single exponential smoothing, the forecast for year 2 is always the (………) of year one?

A

Sales/Data point

8
Q

In relation to significance, a lower alpha…

A

Gives more significance to older data.

9
Q

In relation to significance, a higher alpha gives…

A

Less significance to old data

10
Q

A smaller alpha results in what sort of time-series graph line? And why is this the case?

A

A smoother line.

This is the case because the graph takes longer to react to changes i the data

11
Q

What is the key consideration when setting the alpha?

A

Is the old data significant? If it is we choose a lower alpha, if it is not, we choose a higher alpha.

12
Q

In single exponential smoothing, a line on a graph representing the quick recognition of changes in data points is likely to be more…

A

Erratic

13
Q

Although it is not possible in practise, an alpha of 0 would result in what? And why is this the case?

A

A straight line along the point of the first data entry.

This is because it gives absolute significance to the old data.

14
Q

What is the idea range for an alpha to lie?

A

Between 0.1 and. 0.3

15
Q

What is the Alpha also known as?

A

The smoothing constant

16
Q

Under single exponential smoothing, the weight of all data points ad up to what value?

A

1

17
Q

Under what circumstances is Holts method used?

A

When there is a trend in the data.

18
Q

Under Holts method, 2 equations are needed, what are they?

A

St = Alpha(Y1) + (1-Alpha) (St-1+Bt-1)

Bt = Beta(St - St-1) + (1-Beta)bt-1

19
Q

In the Holt model, what does Beta represent?

A

The smoothing level of the estimated trend.

20
Q

Beta is either given or is calculated how?

A

By taking an estimate from the first 3 data points.

For example: (3rd point - 1st point)/2

21
Q

When making forecasts using the Holts method, how are they calculated?

A

The St of the last period + the Bt of the last period.

22
Q

Between what values can Beta be?

A

0 and 1