w5

Question 1

Forecasting

Answer 1

prediction with time.

Question 2

Cross-sectional

Answer 2

time is not the independent variable.

Question 3

Time-series

Answer 3

past data is used to predict trends.
- usually done with a line chart to show trend
- regular intervals of time (quarters, months, years)

Question 4

Horizontal Pattern

Answer 4

Mean is not changing.

Question 5

Non-stationary time series

Answer 5

There is a jump up or down in the pattern.

Question 6

Trend Pattern

Answer 6

increasing or decreasing over time.

Question 7

Seasonal Pattern

Answer 7

Pattern repeats in regular intervals; repeats across season, quarters, or years.

Question 8

Trend Seasonal Pattern

Answer 8

Has a repeating pattern but also is increasing or decreasing over time.

Question 9

Forecast error

Answer 9

difference between actual value and predict value; like checking after the fact how well your prediction predicted the actual value.

Question 10

Four measures of time series

Answer 10

• Mean forecast error
• Mean absolute error
• Mean square error
• Mean absolute percentage error

Question 11

MFE pros and cons

Answer 11

pros: can be used to determine whether the model is overestimating (negative result) or underestimating (positive result).
cons: negative and positive numbers cancel each other out. Not a good overall measure.

Question 12

MAE pros and cons

Answer 12

pros: solves MFE’s flaw. Better than MSE when you include outliers.
cons: not comparable across different models.

Question 13

MSE pros and cons

Answer 13

pros: most popular overall model measure. Solves MFE’s flaw.
cons: not comparable across different models. Easily biased by outliers.

Question 14

MAPE pros and cons

Answer 14

pros: comparable across models.
cons: most difficult measure to calculate.

Question 15

Non-moving Methods

Answer 15

Naive bar casting method & Average past values casting

Question 16

Naive bar casting method

Answer 16

simply takes the last actual value and uses that value to predict the next value. Adapts to changes quickly making it a good method for non-stationary patterns.

Question 17

Average past values method

Answer 17

average all past actual values to predict the next value. Good for horizontal pattern.

Question 18

Moving methods

Answer 18

Use most recent actual values to predict the next value. They include Moving average forecasting & Exponential Smoothing.

Question 19

Moving Average Method

Answer 19

Uses k number of recent actual values, averages them, and uses that average to predict the next value. Lower K works best with frequent fluctuations while higher K works best with fewer fluctuations.

Question 20

Exponential Smoothing Method

Answer 20

y_t+1 = y(a) * y_hat(1-a); uses the weighted most recent prediction and value to predict then next value. When there is higher variation, a should be higher. When there is less variation, a should be lower. 0 < a < 1.

Question 21

Causal Forecasting

Answer 21

time and a non-time independent variable used to predict dependent variable.

Question 22

Linear Regression for Seasonal Trend Pattern

Answer 22

y = B0 + B1Q1 + B2Q2 + B3Q3 + B4X + e; the time intervals are categorical dummy variables in this instance.

Question 23

How are K and alpha determined?

Answer 23

computers try different values to maximize MSE. Where MSE is maximized, k or alpha is best.