Managerial Economics in a
Global Economy,
Chapter 5
Demand Estimation and Forecasting
Presented by: Saeed
Abdulrahman
Meaning & Importance
• Demand Estimation and Forecasting
• Demand Estimation: finding the current relationship
between quantity demanded and its main factors (price,
income, etc.).
• Demand Forecasting: predicting future demand based
on past data and trends.
• Importance:
• Helps in pricing and production planning
• Useful for inventory control
• Supports marketing and investment decisions
The General Demand Function
• Qd = f (P, Y, Ps, Pc, T, A, N)
• Where:
• P: price of the good
• Y: consumer income
• Ps: price of substitutes
• Pc: price of complements
• T: consumer tastes/preferences
• A: advertising or promotion
• N: number of consumers or population
• These factors together explain how much of a product
people will buy.
Regression Analysis for Estimation
Regression means finding the relationship between two or more variables
usually one dependent variable (Y) and one or more independent variables (X).
Regression shows how a change in X affects Y.
•
Example of Elasticities
•
Qualitative Forecasts
• Survey Techniques
• Planned Plant and Equipment Spending
• Expected Sales and Inventory Changes
• Consumers’ Expenditure Plans
• Opinion Polls
• Business Executives
• Sales Force
• Consumer Intentions
Time-Series Analysis
• Secular Trend
• Long-Run Increase or Decrease in Data
• Cyclical Fluctuations
• Long-Run Cycles of Expansion and Contraction
• Seasonal Variation
• Regularly Occurring Fluctuations
• Irregular or Random Influences
First graph shows the main
patterns in sales over time.
The top one shows a secular trend
the long-term direction and
cyclical fluctuations from
business cycles.
The second bottom shows
seasonal variations that repeat
every year and some random
influences like unexpected events.
These patterns help us understand
and forecast demand more
accurately.”
Trend Projection
• Linear Trend:
St = S 0 + b t
Sₜ = value in year t (like sales or demand), S₀ = starting value (at time 0), b
= growth per period (for example, +5 units every year),
①This means sales increase or decrease by a fixed amount each period.
Example: if sales start at 100 and increase by 5 every year → 100, 105,
110, 115…
Constant Growth Rate (Exponential Trend):
St = S0 (1 + g)t
g = growth rate (for example, +10% each year), Here, sales grow by a
percentage each period (not a fixed number).
Example: if S₀ = 100 and g = 10% → 100, 110, 121, 133.1…
• Estimation of Growth Rate(Taking the natural log (ln) makes the data
easier to analyze:)
lnSt = lnS0 + t ln(1 + g)
• This helps us find the value of g (growth rate) using regression
analysis.
Seasonal Variation
Ratio to Trend Method
Actual
Ratio =
Trend Forecast
1)If the ratio is above 1, actual sales are higher than normal (strong season).
2)If it’s below 1, actual sales are lower (weak season).
Seasonal Average of Ratios for
=
Adjustment Each Seasonal Period
Adjusted Trend Seasonal
Forecast = Forecast Adjustment
Seasonal Variation
Ratio to Trend Method:
Example Calculation for Quarter 1
Trend Forecast for 1996.1 = 11.90 + (0.394)(17) = 18.60
Where 11.90 → the intercept (starting point of the trend line), 0.394 → the slope (how
much the trend increases each period), 17 → the time value (t) for year 1996.1,and So, the
trend value (without seasonal effect) is 18.60
Seasonally Adjusted Forecast for 1996.1 = (18.60)(0.8869) = 16.50
Where 0.8869 is the seasonal adjustment factor for Quarter 1 (Q1), It means sales in Q1 are
usually about 88.69% of the trend level., Therefore, the final forecast that includes seasonality =
16.50.
Year t Trend Forecast Actual Ratio
1992.1 13 12,29 11 0,8950
1993.1 14 13,87 12 0,8652
1994.1 15 15,45 13 0,8414
1995.1 16 17,03 14 0,8221
17
sesonal
adjustemnt for Q1 0,8559
Example
Time 2003.1 2003.2 2003.3 2003.4 2004.1 2004.2 2004.3 2004.4
period
Quantity 11 15 12 14 12 17 13 16
Time 2005.1 2005.2 2005.3 2005.4 2006.1 2006.2 2006.3 2006.4
Period
Quantity 14 18 15 17 15 20 16 19
Moving Average Forecasts
Forecast is the average of data from w
periods prior to the forecast data point.
w
At −i
Ft =
i =1 w
Exponential Smoothing
Forecasts
Forecast is the weighted average of
the forecast and the actual value from
the prior period.
Ft +1 = wAt + (1 − w) Ft
0 w 1
Root Mean Square Error
Measures the Accuracy
of a Forecasting Method
RMSE =
(A − F )
t t
2
n
Moving Average
Three-Quarter & Five Quarter Forecast & Comparison
Quarter Firm’s Three
Actual Quarter MA A-F (A-F)^2
Market Forecast
Share (F)
(A)
1 20 - -
2 22 - -
3 23 - -
4 24 21.67 2.33 5.4289
5 18 23.00 -5
6 23
7 19
8 17
9 22
10 23
11 18
12 23 21.0
Total 78.3534
13 21.33 20.6
Exponential Smoothing
Quarter Firm’s Forecast Forecast
Actual with w = A-F (A-F)^2 with A-F (A-F)^2
Market 0.3 w=0.5
Share (F) (F)
(A)
1 20 21 -1.0 1.0
2 22 20.7 1.3 1.7
3 23 21.1 1.9 3.6
4 24 21.7 2.3 5.5
5 18 22.4 -4.4 19.0
6 23 21.1 1.9 3.8
7 19 21.6 -2.6 7.0
8 17 20.8 -3.8 14.8
9 22 19.7 2.3 5.3
10 23 20.4 2.6 6.8
11 18 21.2 -3.2 10.0
12 23 20.2 2.8 7.7
13 21.1 Total 86.3
RMSE 2.682215
Barometric Methods
• National Bureau of Economic Research
• Department of Commerce
• Leading Indicators
• Lagging Indicators
• Coincident Indicators
Econometric Models
Single Equation Model of the
Demand For Cereal (Good X)
QX = a0 + a1PX + a2Y + a3N + a4PS + a5PC + a6A + e
QX = Quantity of X PS = Price of Muffins
PX = Price of Good X PC = Price of Milk
Y = Consumer Income A = Advertising
N = Size of Population e = Random Error
Applying the data
Thank You