term1 Definition1term2 Definition2term3 Definition3
Please sign in to your Google account to access your documents:
Linear Equation
Y= mX + b
Y= dependent variable
X= independent variable
m = slope (tells us about the direction of relationship)
b = intercept
Linear equation lets us perform the following tasks:
-summarize the relationship among variables by an equation
-make predictions on a variable based on data for related variables (Thing 1)
-Measure the sensitivity of one variable to changes in a related variable (Thing 2)
Regression equation
Ŷ= b0+ b1X
- Ŷ = dependent variable, or response or predicted variable
- X is the independent variable, or predictor or explanatory variable
- Simple regression, only 1 predictor variable (X).
- In multiple regression, 2 or more X, denoted as:X1, X2, X3, …, Xn
- b0 is the intercept of the regression line
- b1 is the slope of the regression line
Prediction and Forecasting
Thing 1: Plug and Chug
- Dep var = intercept + slope * Indep var
- Predicted Y = Y-hat = Ŷ= b0+ b1X
Given a value for X, plug into the regression equation and chug out the value for Y
Example:firstclass mail volume (1stClVol) and population (PopUSA)
¤Predicted (1stClVol) = - 116 + 0.80 * (PopUSA)So, tell me population and I’llpredict or forecast first class mail volume.
Example: if population is expectedto be 370: ¤Predicted1stClVol = - 116 + 0.80*(370)= 180
Thing 2: Marginal Effects
- Predicted Y = Ŷ= b0+ b1X
- Given a change in X, the change in Y is the slope times the change in X
changein Y = slope * change in X
∆Y= b1* ∆X
¤Predicted (1stClVol) = - 116 + 0.8*(PopUSA)So, tell me the change in populationand I’ll predict the change in first class mail volume. So, if population isexpected to grow by 10 million then mail volume is expected to increase by 8billion:
∆Y= slope*∆X = 0.8*(10)= 8
Time Series
Special Case:
When time itself is the independent variable in the simple regression analysis, the equation is:
Predicted Y = Ŷ= b0 + b1 Time
Time trends: upward, downward or no trend
Trend Rates only apply to time trends
The trend rate is the coefficient of the time variable, or the slope of the regression equation (b1)
Special Case Example:
Example of unemployment:
Un = Unemployment in percent of thelabor force
The regression equation is the timetrend:
Ŷ = Predicted Un% = b0 + b1Time
Result: Predicted Un% = 10.13 - 0.06Month
Thing1: What will unemployment be in six months?
Predicted Un% = 10.13 - 0.06*(37) = 7.9%
Thing 2: Unemployment falls by 0.06 (b1)percentage pts/month, 0.18 pts/quarter,or 0.72 pts/year
Need help typing ? See our FAQ (opens in new window)
Please sign in to create this set. We'll bring you back here when you are done.
Discard Changes Sign in
Please sign in to add to folders.
Sign in
Don't have an account? Sign Up »
You have created 2 folders. Please upgrade to Cram Premium to create hundreds of folders!