QM
Measures of Relationships and
Forecasting (Trends)
By
Kaconco James
0772653191
1
Introduction
• Regression is the measure of the average
relationship between two or more variables (in
terms of the original units of the data).
– Independent (explanatory) variable (x)
• Time : time series
• Not time: causal (associative) relationship
– Dependent (explained) variable (y)
• The mathematical relationship developed can
be used for predicting (forecasting)
2
Regression Types
• Simple regression:
– One independent variable
– One dependent variable
– Example of regression equation: y = a + bx
• Multiple regression
– Deals with more than one independent variable
– Example of multiple regression equation:
y = a +b1x1 + b2x2 + …
01/22/25 [email protected] 3
Computation of Regression Coefficients
• The regression equation: y = a + bx
• Where,
y is the dependent variable
x is the independent variable
a and b are regression coefficients
The direct method will be used to determine
the values of a and b
4
Direct Method
5
Example: Regression
Example 1 Example 2
Period X (Time) Y (sales X (wins) Y
(Years) in units) (Attendance)
2001 1 8 4 36
2002 2 14 6 40
2003 3 12 6 41
2004 4 16 8 53
Forecasting
Example 1:
Determine the sales when x = 5. What is the x (time) value that corresponds to a
sale of 30 units?.
Example 2:
What attendance corresponds to a win of 5 games and what win corresponds to an
attendance of 100?
6
Regression Summary
• It studies the nature of relationship between
variables (independent and dependent).
• It is basically used for predicting (forecasting).
• It is used to study the cause and effect
relationship.
7
Correlation
• Helps to determine the degree of relationship
that exists between two or more variables:
– Direction (positive or negative).
– Magnitude (perfect, strong, moderate, weak
positive or negative or no relationship).
• Benefits
1.Find a single figure to measure relationship.
2.Reduces uncertainty in decision making.
3.Can be used to study regression (relationship).
8
Correlation Coefficient (r)
– A numerical value (direction, magnitude) determined to express the
relationship that exists between two or more variables
– Measures the direction (positive, negative, zero) and
– Measures strength (magnitude) is between the range -1 to +1.
• The closer the r value is to +1.0 the better the regression line fits the data points.
• r is positive if Y increases as X increases, and
• r is negative if Y decreases as X increases.
• r of zero indicates an absence of any relationship between the two variables.
Computing or Measuring Correlation
Coefficient
• Direct method referred to as Pearson
Correlation Coefficient
• The equation:
• Examples: continue the computation with the
previous correlation problems
10
Coefficient of determination (r2)
• It is the square of coefficient of correlation
• It is the ratio of the explained variance to total variance
• Example: suppose the value of r = 0.9,
– then r2 = (0.9)2 = 0.81 or 81%
• This means that 81% of the variation in the
dependent variable has been explained by
(determined by) the independent variable(s).
– The 19% is due to other factors, not considered in the
study
– The other factors are called coefficient of non-
determination (K2) = 1 – r2
11
Merits of Coefficient of Correlation
1. This is the most widely used algebraic method to measure
coefficient of correlation.
2. It gives a numerical value to express the relationship
between variables.
3. It gives both direction and magnitude (degree) of
relationship between variables.
4. It can be used for further algebraic treatment such as
coefficient of determination and coefficient of non-
determination etc.
5. It gives a single figure to explain the accurate degree of
correlation between two variables.
12
Demerits of Coefficient of Correlation
1. It is very difficult to compute the value of
coefficient of correlation.
2. It is very difficult to understand.
3. It requires complicated mathematical calculations.
4. It takes more time.
5. It is unduly (excessively) affected by extreme items.
6. It assumes a linear relationship between the
variables. But in real life situation, it may not be so.
13
