Forecasting Test 2: CH.4,5,6

What is a major limitation of all regression techniques? provide an example

Only suggests relationship doesn't give a cause. Relation does not mean causation. Ex: Murders and Ice cream Both increase when its warm out but unrelated lets not assume murders caused ice cream sales to increase.

Before considering other statistical diagnostics, the following is (are) initial evaluative steps for bivariate regression models

All of the above

Visual inspection of the data will help the forecaster identify

All of the above

Can we easily account for these seasonal patterns using the typical casual variables that we use regression analysis?

No we need to use dummy variables

A seasonal factor greater than 1 indicates a period in which Y is less than the yearly average.

False

Business and economic data series never display seasonal patterns that recur with some regularity year after year.

False

Intercept and Slope are terms associated with the dependent variables in multiple regression models.

False

Smoothing constant analysis is a statistical tool that gives the ability to estimate the mathematical relationship between a dependent variable and a single independent variable.

False

The index of leading economic indicators is not one of the three noteworthy possible business cycle indicators.

False

Time-series decomposition models are popular because they are difficult to understand and explain.

False

Two data sets that have the exact same estimated linear regression must be the same data.

False

What advice would you give someone who was charged w/ developing forecasts with regression models and why would you give that advice?

KIS- keep it simple Harder to communicate, take more time, more costly otherwise

Which of the following is a method that can be used to evaluate forecast results?

RMSE

The process of deseasonalizing data has the following useful results:

A&B Above

How can dummy variables help when trying to account for pronounced seasonality or any other qualitative attribute? Also, describe how dummy variables work?

-gives us a method to incorporate qualitative variables -0 or 1 if condition does not exist=0 condition does exist=1

You are the sales manager for Tom's Turkey Farms, which is a major producer of turkey food products. Discuss at least 3 reasons why you might prefer to use time-series decomposition to generate sales forecasts.

-Adjust for trends/ see trends like seasonality ex: Thanksgiving -Time series is easy to understand -Time series info is consistent with the way Managers look at data. -easier to look at each component -accurate forecast -Forecast for month or quarter

What are the advantages in showing data in graphic form rather than, or in addition to, tabular form?

-Can be easier to process or interpret; more straight forward -Trends are easier to see/outliers as well -Can see slope and intercept

Discuss the steps you should particularly focus on in a process for applying regression forecasting

-Data considerations (trends, seasonality, identifying variables) -Model Selection -Model Evaluation (MAPE) -Forecast Prep

Which of the following is not a reason to employ simple linear regression to generate sales forecasts for a retail outlet store?

None of the Above

Which of the following is not a recommended step in preparing a forecast using the simple linear regression model?

None of the above (Because they all are used in preparing a forecast using the simple linear regression model).

Write a description of each of the 4 components of the classic time series decomposition technique.

T = long-term trend based on deseasonalized data Often called the centered moving-average trend (CMAT) Since deseasonalized data are centered moving averages (CMA) of the original Y values S = seasonal indices (SI) Normalized averages of seasonal factors that are determined as ratio of each period's actual value (Y) to the deseasonalized value (CMA) for that period C = cycle component Cycle factor (CF) is ratio of CMA to CMAT Represents gradual wavelike movements in series around trend line I = irregular component Assumed equal to 1 unless forecaster has reason to believe a shock may take place Then I could be different from 1 for all or part of the forecast period

The Y-intercept of a regression line is -14 and the slope is 3.5. Which of the following is correct?

The regression line crosses the Y-axis at -14.

What is the general purpose of multiple regression analysis? Give two examples include dependent variable and more than 1 independent variable

To learn more about the relationship between several independent or predictor variables and a dependent variable Examples: Test scores =Dependent What did students study= independent Income of Families= independent Time of day of the Test = independent Salary= d Education level= I Past Experience = I Time @ company = I

Bivariate regression analysis is also referred to as simple linear least-squares regression.

True

Business and economic data used in forecasting are most often time-series data.

True

Classical time-series decomposition involves the ratio-to-moving-average technique.

True

In developing forecasts with regression models, perhaps the best advice is to follow the "KIS" principle: keep it simple.

True

Many business and economic time series contain underlying components that, when examined individually, can help the forecaster better understand data movements and therefore make better forecasts.

True

Multiple regression is a statistical procedure in which a dependent variable (Y) is modeled as a function of more than one independent variable (X 1, X 2, X 3, ...., X n).

True

The cyclical component of a time series is measured by a cycle factor (CF).

True

The cyclical component of a time series is the extended wavelike movement about the long-term trend.

True

The first thing you should do in reviewing regression results is to see whether the signs on the coefficients make sense.

True

The ordinary least-squares (OLS) criterion for the best multiple regression model is that the sum of the squares of all the error terms be minimized.

True

There was a time when regression lines were estimated in an ad hoc manner, based solely on an analyst's visual interpretation of the data.

True

When evaluating alternative multiple regression models, it is better to compare Root-Mean-Squared-Errors (RMSE) than R-squared values.

True

The classical time-series decomposition forecasting model can be represented by the following simple algebraic statement:

Y = T x S x C x I

In using regression analyses we begin by supposing that Y is a function of X. That is

Y = f(X)

Can data series display pronounced seasonal patterns that recur with regularity? Give two examples of patterns and types of forecasts these affect?

Yes -Chocolate sales are affected by seasonality in Oct, Dec, and Feb = seasonal patterns Holidays -Spending in April and December B/c Christmas and Tax returns. -Flower Sales Feb and May -Bathing suit sales in summer months

The sign on the slope estimate in a regression problem

a. always has the same sign as the correlation of Y and X.

The following type of a variable may be qualitative or nominally scaled (such as the season of the year). These variables typically take on a value of either 0 or 1. a. dummy variables b. replica variables c. copy variables d. imitation variables

a. dummy variables

Time series contain _________ components.

all of the above

The values of sample statistics for determining intercept (b 0) and slopes (b 1, b 2, .... b n) are almost always determined by utilizing: a. a table in a textbook. b. a table in government regulation. c. a computer software package. d. none of the above.

c. a computer software package.

____________ are often found in most regression output. a. Standard errors b. Sample size (number of observations) c. Adjusted R Square d. all of the above

d. all of the above

Sample regression model forecast errors (deviations of predicted values from actual values of Y) are called

residuals

Income is used to predict savings. For the regression equation Y = 1,000 + .10X, what is Y?

savings

A model is said to be __________ if the model is incomplete and therefore more than one independent variable is needed.

underspecified