Business Forecasting Topic 5 Flashcards

1
Q

extending simple exponential smoothing formulae

A

latest estimate = weighted average of 2 estimates of level
1. latest observation
2. previous estimate of level

larger α = more weight on more recent = more responsiveness to new situations

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

inaccuracy for SES

A

not accurate forecasting method when upward or downward trend or seasonal pattern

  • forecasts tended to lag behind pattern
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Holt’s method

A
  • extension of SES
  • produce forecasts where linear trend which is subject to changes
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Holt method updates estimates of 2 values

A

after new observation

  1. underlying level of series at that point in time (changes period by period)
  2. the trend
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

trend

A

at a point in time
difference between levels in consecutive periods

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

Holt’s method of updating estimate of underlying level

A
  • identical to SES but there is a trend in series - previous level not good guide for current level
  • trend = expect level to change dont want previous level estimate on our weighted average -> overcome this by add our latest trend estimate to last period’s estimate of level
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Holt’s method of updating the estimate of the trend

A

FORMULAE SHEET!

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

Holt’s method smoothing constant

A

different smoothing constant for the trend
β
between 0 and 1

  • allows to have different degrees of responsiveness to changing levels and changing trends - if time series suggest this is appropriate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

making a forecast for Holt’s method

A

FORMULAE SHEET!
- - add our latest trend estimate to latest level estimate
assume trend is linear = make forecasts for longer periods ahead

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Holt’s method formulae summary

A

FORMULAE SHEET!

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

Holt’s method starting values

A

need initial for level and trend

LEVEL - set level equal to first observation
L1 - Y1

TREND - let initial trend estimate = difference between first 2 observations
b1= Y2-Y1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

optimal α and β values

A

chosen to minimise MSE or some other measure

Excel solver used here

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

Holt’s method

A
  • assume linear trend (increase at current rate)
  • form of Holts = Damped Holts = assume future increase rate will gradually decline = damped trend projected
  • method not used for seasonal data (unless deseasonalised)
  • if α = β -> Brown’s double exponential smoothing (holt = more flexible than brown as different levels of responsiveness to changes on level and trend
  • can handle horizontal pattern
  • assign initial trend as zero and beta is 0 this becomes the SES method

underlying level = forecast if beta is zero

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

Holt-Winters method

A
  • series have both trend and seasonal pattern both subject to change over time
  • only handling multiplicative seasonality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Holt Winters updates estimates

A

new observations available:

  1. underlying level of sales
  2. underlying trend
  3. seasonal pattern
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

underpinning

A
  1. seasonal index -> ration of actual observation to height of underlying trend line - height represented by underlying level of series
    = actual at time t divided by level at time t
  2. deseasonalised value = actual value divided by appropriate average seasonal index
17
Q

Holt winters method of updating the estimate of the level

A
  • use deseasonalised latest observation
18
Q

Holt Winters updating the estimate of the trend

A

same as Holt method

19
Q

Holt Winters updating the estimate of the seasonal index for current season

A
  • third smoothing constant ɣ
20
Q

seasonal index smoothing constant for Holt Winters

A

ɣ gamma
between 0-1

closer to 1 indicates differing seasonal deviations across the cycles in the data

21
Q

making a forecast for Holt Winters method

A

FORMUALE SHEET

22
Q

Holt winters formulae

A

FORMULAE SHEET!

23
Q

Holt Winters starting level value

A

LEVEL - average of monthly or quarterly observations for first year

24
Q

Holt winters starting trend value

A

crude method -
monthly data = actual in yr 2 - actual in yr 1 divided by 12

quarterly data = actual in Q1 year 2 - actual in Q1 of year 1 divided by 4

25
Q

Holt Winters starting seasonal index

A

actual for January or Q1 divided by starting level

26
Q

holt winters optimal values α β ɣ

A

chosen to minimise MSE or other measure

Excel Solver

27
Q

Holt’s Winters method

A
  • assume trend is linear - assume increasing at current rate
  • where think trend is changing but seasonal pattern is stable -> prefer to use Holt method on deseasonalised data then re-seasonalise to obtain forecast
28
Q

impact of initial values

A

substantial impact deepens on values of alpha beta gamma and the length of the data series

29
Q

linear trend
inherent randomness
stable seasonal pattern

A

holt method using deseasonalised data