You are on page 1of 12

Chapter 7: LINEAR REGRESSION

Linear Regression

 Approximating the relationship between the dependent variable and independent variable by a
straight line.

 Uses a two-dimensional graph with sales measured along the vertical axis and time measured
along the horizontal axis.

 After plotting the sales data, this method finds a line passing through the midst of the data as
closely as possible.

 The extension of the line into future months provides the forecast of sales in these future
months.

 linear regression line – the line that approximates the relationship between the dependent and
independent variable when using casual forecasting.

 independent variable - a quantity that drives the value of the independent variable in casual
forecasting. it is called explanatory variable

 dependent variable – the quantity of interest when doing a casual forecasting (eto yung
hinahanap or ifoforecast) or predictor variable

 formula for linear regression line Is y=a+bx

o where y is the dependent variable

o a is the intercept

o b is the slope

o x is independent variable

 R square – the closer to 1, the better the regression line fits the data.

 to check if results are reliable, look at Significance F,if this value is less than 0.05, you’re OK. if
Significance F is greater than .05 better stop using the set of independent variables . Delete a
variable with a high P-value (greater than 0.05).

 Most or all P-values should be below 0.05.

Step 1: using SLOPE and INTERCEPT function

Intercept function arguments are the range of dependent and independent variable
Slope function – arguments are the range of dependent variable and independent variable

 to get the slope value of slope, you have to use the range values of depenent and independent
variable

 to get the intercept , again you need to use the range values of dependent and independent
variable
 Then to solve or get the forecasted value you have to apply the formula y=a+bx. substitute the
values.

 Another way is using the TREND function.


 trend – is a function that returns numbers in a linera trend matching known data points, using
the least squares methods


 Another way is using the Analysis tool pak.
o go to Data tab

o then in the analysis group, select Data Analysis


o in the data Analysis Tools, Select Regression, and click ok

o regression will apprear,

o in input Y range, you have to select the range of your dependent variable, (if you will

include the labels, tick the Labels )


o in input X range, select your independent variables
o then if you want to display the summary table in the same sheet, choose Output Range,
select the specific cecll reference on where you want it to display.
o then to compute the forecast value, you have to look and use the values in coefficients

column
o apply the formula y=a+bx, if you have multiple independent variables, use this formula,
y=a+b1x1+b2x2…bnxn, where b1 is the coefficient of 1 independent variable, b 2 is the
coefficient of the second independent variable.
o
o

Last-Value Forecasting Method

 A method that uses the last value of a time series as the forecast of the next value
 The last-value forecasting method ignores all data points in a time series except the last one.

Forecast = Last value


Step 1: create a model


Step 2: create a forecast value (apply the formula)

*copy the formula up to the last given value of actual or true value.

*ISNUMBER() – is a function that checks whether a value is a number and returns a true or false

*NA() – is a function that returns #N/A (value not available)

Step 3: Forecasting Error


*copy the formula up to E12

*ABS – a function in excel that returns the absolute value of a number without its sign.

Step 4: Compute for mean absolute deviation (MAD)


Step 5: Compute for Mean Square Error
Other way is to manually compute it by adding another column for Error 2

Step 6: Create a chart

select values for true value and last value forecast


*click insert tab

*select line chart and choose Line with markers


*to change the legend click on one data point

*in the formula bar add the input the new legend

You might also like