Exam 3 QMB3250 UF
Smoother - ANSFor Simple moving average, longer periods are _________ than shorter
periods.
Smoother or less smooth
Simple Moving Average - ANSCreates a new time series from the original data by averaging
adjacent values.
helps illuminate how the data is varying over time.
Affected by outliers and peaks.
Doesn't work well when there are strong seasonal, trend or cyclical components.
Simple exponential smoothing - ANSvalues that are closest to the estimated value get the most
weight in the moving average and values that are farther away get less.
adjust weight between recent and historic data.
Smoother - ANSA smaller a (alpha) would be __________________.
Smoother or more flat or less smooth
Autoregressive models - ANSTime series where we add in columns containing lags.
Good for quarterly lag.
Multiple Regression Models - ANSWe can cautiously predict a little further away from the data.
Better at fitting series that have trend, cyclical, and seasonal components.
Linear
- Additive: SLR
- Exponential: SLR but log(y)
Quarters: Add indicator variables
Random walk - ANSthe best prediction is the most recent point. Time series.
Sometimes described as white noise.
One way ANOVA - ANSAllow us to compare two or more means.
Ho: M1 = M2 = Mk (No difference in treatments)
Ha: At least one population mean is different from the rest.
exogenous variable - ANSan extra variable added to help predict this cyclical trend.
sum of squares treatment - ANSestimates how much the means vary.
gets bigger by adding another group.
sums of squares error - ANSestimates how much variation there is within each group.
Smoother - ANSFor Simple moving average, longer periods are _________ than shorter
periods.
Smoother or less smooth
Simple Moving Average - ANSCreates a new time series from the original data by averaging
adjacent values.
helps illuminate how the data is varying over time.
Affected by outliers and peaks.
Doesn't work well when there are strong seasonal, trend or cyclical components.
Simple exponential smoothing - ANSvalues that are closest to the estimated value get the most
weight in the moving average and values that are farther away get less.
adjust weight between recent and historic data.
Smoother - ANSA smaller a (alpha) would be __________________.
Smoother or more flat or less smooth
Autoregressive models - ANSTime series where we add in columns containing lags.
Good for quarterly lag.
Multiple Regression Models - ANSWe can cautiously predict a little further away from the data.
Better at fitting series that have trend, cyclical, and seasonal components.
Linear
- Additive: SLR
- Exponential: SLR but log(y)
Quarters: Add indicator variables
Random walk - ANSthe best prediction is the most recent point. Time series.
Sometimes described as white noise.
One way ANOVA - ANSAllow us to compare two or more means.
Ho: M1 = M2 = Mk (No difference in treatments)
Ha: At least one population mean is different from the rest.
exogenous variable - ANSan extra variable added to help predict this cyclical trend.
sum of squares treatment - ANSestimates how much the means vary.
gets bigger by adding another group.
sums of squares error - ANSestimates how much variation there is within each group.
