Forecasting 1, del 1

Question 1

What are quantitative methods based on?

Answer 1

Quantitative methods are based on an analysis of historical data concerning one or more time series

Question 2

Vad är casual method?

Answer 2

Casual method is when the historical data used involve other time series that are believed to be related to the time series that we are trying to forecast

Question 3

Vad är a time series?

Answer 3

Set of observations measured at successive points in time or over successive periods of time. (Timme för timme eller varje månad)

Question 4

Vad är time series method?

Answer 4

Time series method is when the historical data used are restricted to past values of the series that we are trying to forecast

Question 5

Exempel på time series method:

Answer 5

Om vi vill förutsäga framtida försäljning av en produkt så tittar vi bara på tidigare försäljningssiffror

Question 6

3 olika time series methods

Answer 6

Smoothing Methods
Trend projection
Trend projection adjusted for seasonal influence

Question 7

Vad är smoothing methods?

Answer 7

Smoothing Methods used in cases which the time series is: fairly stable, has no significant trend, seasonal, or cyclical effects.

This to average out the irregular components of the time series. Används för att skapa jämna prognoser och jämna ut sluppmässiga fluktuationer i data

Question 8

Fyra olika sorters smoothing methods:

Answer 8

Moving averages
Centered moving averages
Weighted moving averages
Exponential smoothing

Question 9

Förklara Moving averages

Answer 9

Computing an average of the most recent n data values for the series and using this average for forecasting next period

Question 10

Förklara centered moving averages

Answer 10

Computing an average of n periods data and associating it with the midpoint of the periods. (Average for period 5, 6, 7 is associated with period 6) Useful in the process of computing season indexes

Question 11

Förklara weighted moving averages

Answer 11

Computing the average of the most recent n periods. They are typically given more weight than older observations.

Question 12

Förklara exponential smoothing

Answer 12

Computing forecast for the next period that is equal to the forecast for the current period + a proportion of the forecast error in the current period (a)