1. Forecasting
1.1. Seasonality
You need to differentiate between seasonal products (season being Christmas) and non-seasonal
products.
For seasonal products it is important to incorporate the data from previous seasons (long
time-period)
For non-seasonal products it might be more reliable to incorporate more recent historical
data
1.2. Trend
If you only take the data from last year december to create your forecast. You do not take any trends
into account. For example;
Growing number of stores
Increasing demand for vegan products
Recent global events
1.3. Causal effects
Causal effects to take into account:
Promotions (from Jumbo or producer)
Regional or national media
Others;
Out of stocks (bullwhip effect)
Level of aggregation
,1.4. Demand Patterns
Time series
The repeated observations of demand for a service or product in their order of occurrence
There are five basic time series patterns
Horizontal
Trend
Seasonal
Cyclical
Random
Horizontal: Data cluster about a horizontal line
Trend: Data consistently increase or decrease
Seasonal: Data consistently show peaks and valleys
Cyclical: Data reveal gradual increases and decreases over extended periods
,2. Type of forecast techniques
Judgement methods - Qualitative
Causal methods - Quantitative
Time-series analysis - Quantitative
Trend projection using regression - Quantitative
2.1. Causal methods
Dependent variable – The variable that one wants to forecast
Independent variable – The variable that is assumed to affect the dependent variable and thereby
“cause” the results observed in the past
Simple linear regression model is a straight line
Y = a + bX
where
Y = dependent variable (e.g. turnover)
X = independent variable (e.g. marketingbudget)
a = Y-intercept of the line (y-axis)
b = slope of the line
2.1.1. Correlation analysis (Regression)
2.1.2. The scatterplot
2.1.3. Correlation analysis
Y = a + bX
where
Y = dependent variable (e.g. turnover)
X = independent variable (e.g. marketingbudget)
a = Y-intercept of the line (y-axis)
b = slope of the line
, 2.2. Time Series Methods
Naïve forecast
The forecast for the next period equals the demand for the current period (Forecast = Dt)
Horizontal Patterns: Estimating the average
Simple moving average
Weighted moving average
Exponential smoothing
2.2.1. Simple Moving Averages
Specifically, the forecast for period t + 1 can be calculated at the end of period t (after the actual
demand for period t is known) as
where
Dt = actual demand in period t
n = total number of periods in the average
Ft+1 = forecast for period t + 1
2.2.2. Weighted Moving Averages
In the weighted moving average method, each historical demand in the average can have its own
weight, provided that the sum of the weights equals 1.0.
The average is obtained by multiplying the weight of each period by the actual demand for that
period, and then adding the products together.
2.2.3. Exponential smoothing
A sophisticated weighted moving average that calculates the average of a time series by implicitly
giving recent demands more weight than earlier demands
Requires only three items of data
The last periodÕs forecast
The actual demand for this period
A smoothing parameter, alpha (α),
The equation for the forecast is
2.2.4. Choosing the right Time-Series method
For any forecasting method, it is important to measure the accuracy of its forecasts.
Forecast error is simply the difference found by subtracting the forecast from actual demand for a
given period, or
Et = Dt – Ft
where
Et = forecast error for period t
Dt = actual demand in period t
Ft = forecast for period t
1.1. Seasonality
You need to differentiate between seasonal products (season being Christmas) and non-seasonal
products.
For seasonal products it is important to incorporate the data from previous seasons (long
time-period)
For non-seasonal products it might be more reliable to incorporate more recent historical
data
1.2. Trend
If you only take the data from last year december to create your forecast. You do not take any trends
into account. For example;
Growing number of stores
Increasing demand for vegan products
Recent global events
1.3. Causal effects
Causal effects to take into account:
Promotions (from Jumbo or producer)
Regional or national media
Others;
Out of stocks (bullwhip effect)
Level of aggregation
,1.4. Demand Patterns
Time series
The repeated observations of demand for a service or product in their order of occurrence
There are five basic time series patterns
Horizontal
Trend
Seasonal
Cyclical
Random
Horizontal: Data cluster about a horizontal line
Trend: Data consistently increase or decrease
Seasonal: Data consistently show peaks and valleys
Cyclical: Data reveal gradual increases and decreases over extended periods
,2. Type of forecast techniques
Judgement methods - Qualitative
Causal methods - Quantitative
Time-series analysis - Quantitative
Trend projection using regression - Quantitative
2.1. Causal methods
Dependent variable – The variable that one wants to forecast
Independent variable – The variable that is assumed to affect the dependent variable and thereby
“cause” the results observed in the past
Simple linear regression model is a straight line
Y = a + bX
where
Y = dependent variable (e.g. turnover)
X = independent variable (e.g. marketingbudget)
a = Y-intercept of the line (y-axis)
b = slope of the line
2.1.1. Correlation analysis (Regression)
2.1.2. The scatterplot
2.1.3. Correlation analysis
Y = a + bX
where
Y = dependent variable (e.g. turnover)
X = independent variable (e.g. marketingbudget)
a = Y-intercept of the line (y-axis)
b = slope of the line
, 2.2. Time Series Methods
Naïve forecast
The forecast for the next period equals the demand for the current period (Forecast = Dt)
Horizontal Patterns: Estimating the average
Simple moving average
Weighted moving average
Exponential smoothing
2.2.1. Simple Moving Averages
Specifically, the forecast for period t + 1 can be calculated at the end of period t (after the actual
demand for period t is known) as
where
Dt = actual demand in period t
n = total number of periods in the average
Ft+1 = forecast for period t + 1
2.2.2. Weighted Moving Averages
In the weighted moving average method, each historical demand in the average can have its own
weight, provided that the sum of the weights equals 1.0.
The average is obtained by multiplying the weight of each period by the actual demand for that
period, and then adding the products together.
2.2.3. Exponential smoothing
A sophisticated weighted moving average that calculates the average of a time series by implicitly
giving recent demands more weight than earlier demands
Requires only three items of data
The last periodÕs forecast
The actual demand for this period
A smoothing parameter, alpha (α),
The equation for the forecast is
2.2.4. Choosing the right Time-Series method
For any forecasting method, it is important to measure the accuracy of its forecasts.
Forecast error is simply the difference found by subtracting the forecast from actual demand for a
given period, or
Et = Dt – Ft
where
Et = forecast error for period t
Dt = actual demand in period t
Ft = forecast for period t
