Exam 2 Flashcards
(23 cards)
What can regression do?
Testing theory and identifying relationships, measuring strength of a relationship, forecasting
Ordinary least square
Find the line that best fits the data points. Min the sum of the squared distances from individual observations to this line
R2 correlation
Measures variance explained
F test correlation
Test the significance of the overall model to see whether it has explanatory power
T test correlation
Test specfic betas to see wheater that particular indenpendent variable has a linear relationship with the dependent varable
Multicollinearity exists when
Two or more of the independent variables used in regression are correlated
Why do we care about multicollinearity?
Reduces MSR and increases MSE so it decreases fcal and a low fcal fails to reject the null
Standardized coefficient
The absolute value indicates importance of predictors
Dummy variable
A variable that indicates the presence or absence of some characteristics or attributes
Time series analysis
Is a set of quantitative methods for determining patterns in time series data
Trend
Steady tendency of increase or decrease over time. Systematic
Seasonal variation
Regular fluctuations or periodic changes that repeat year after year. Repeats every year. Systematic
Cyclical variation
Repetitive fluctuations or swings of varying length and intensity in the long term. Periods longer than one year. Systematic
Random or irregular varaition
Unpredictable random variations in the time series that the above three components fail to account for. Short non repeating unsystematic random
What does the moving average do
Allows us to eliminate or smooth out the fluctuations in the time series data
Seasonal index
Expresses the value of a time series variable for each period as a percentage of the trend value or moving average for that time period
Parametric test
Inferences based on assumptions about the nature of the population distribution
Nonparamentric tests
Distribution free methods making no assumptions about the population distribution
Purpose of the goodness of fit test
Determine how well an observed set of data fits an expected outcome
Chi squared
Two categorical variables
Correlation
No iv or dv distiction both continuous
ANOVA
Iv: categorical
Dv: continuous
Regression
Iv: either
Dv: continous