3. The Least error square method is the best fitting technique to fit the data in Logistic Regression. 4. Multiplicative Holt-Winters Method is used if the trend line grows exponentially. 5. In an exponential smoothing model, the higher the smoothing parameters is, the more we give a weight to the recent observations. 6. The value of covariance of stationary time series, xt , is dependent on time t. 7. If xt is a stationary process, then the first difference of xt is also a stationary process.
Answer each question TRUE or FALSE. If FALSE give reason.
1. Modeling the logistic regression, ROC (receiver operating characteristic) curve is used to determine the best cutoff value to use for forecasting.
2. The cutoff value with the highest accuracy in the logistic regression is the best cutoff value in general. However, in some cases, sensitivity or specificity can be our most concern.
3. The Least error square method is the best fitting technique to fit the data in Logistic Regression.
4. Multiplicative Holt-Winters Method is used if the trend line grows exponentially.
5. In an exponential smoothing model, the higher the smoothing parameters is, the more we give a weight to the recent observations.
6. The value of
7. If xt is a stationary process, then the first difference of xt is also a stationary process.
8.The third regular differences of the time series values y1, y2, ..., yn are zt = yt − 2yt−1 + 2yt−2 − yt−3.
9. If PACF spikes at lag 1 and 2 (with cut off after that) and ACF dies down, the corresponding model is xt = δ + ϕ1xt−1 + ϕ2xt−2 + wt .
10. The following model xt = .4xt−1 + .21xt−2 + wt + .6wt−1 + .09wt−2. is causal and invertible.
.
Step by step
Solved in 5 steps with 5 images
