Statistical Models Used in the Forecasting of Automobile Sales

1 / 15
About This Presentation
Title:

Statistical Models Used in the Forecasting of Automobile Sales

Description:

Statistical Models Used in the Forecasting of Automobile Sales Ben Nelson, Ling-Chih Chen, Hsiu-Jung Hu, ... Manufacturing of rubber & plastic products ... –

Number of Views:179
Avg rating:3.0/5.0
Slides: 16
Provided by: BenandSar
Category:

less

Transcript and Presenter's Notes

Title: Statistical Models Used in the Forecasting of Automobile Sales


1
Statistical Models Used in the Forecasting of
Automobile Sales
  • Ben Nelson, Ling-Chih Chen, Hsiu-Jung Hu,
  • Yuliy Nesterenko, Bin Shi, Kimberly Williams

2
Introduction
3
Objective
The objective of this project was to select the
most viable statistical model to forecast
Auto-sales for the USA. Our exercise included
model building, model validating and model
selection.
4
Models Used
5
Multiple Regression Variables
- Auto Sales
- Number of unemployed persons (thousands) - Bank
credit - Real personal income (billions of
chanied) - Federal funds rate - Japan (yen per
US) - Germany (Deutsche mark per US) -
Manufacturing of autos light trucks -
Manufacturing of rubber plastic products -
Petroleum products consumption - Truck tonnage
index
6
Multiple Regression Variables
- Central South America nuclear electric -
Texas marketed production of natural gas (unit
10,000) - Imports on machinery (transportation
equipment) - Exports on manufactured goods - SP
stock price index on transportation - Producer
price index in finished consumer goods - Real
wages salaries in mining manufacturing -
Expenditures on furniture household equipment -
SPs stock prices (500 common stocks) - Index of
help wanted advertising - Passenger fares
7
Independent Variable Assessment
We transformed 2 variables (Producer price index
of finished consumer goods SP's stock prices
of 500 common stocks ) by using their squared
value.
8
Independent Variable Assessment
Model 1 Auto-sales with variable 14 Exports
of manufactured goods only. Model 2
Auto-sales with variables 13 and 14 Exports
of manufactured goods and Imports of machinery
transportation equity. Model 3 Auto-sales
with variables 14, 2, and 19 Exports of
manufactured goods, Bank credit and
Expenditures on furniture household equity.
9
Multiple Regression Analysis
Model 1
Model 2
Model 3
Probability
0.000000
0.000000
0.000000
Adj R-Squared
0.6517
0.6529
0.6479
Significance component variables
1 of 1
1 of 2
1 of 3
Linearity
OK
OK
OK
Independence
OK
OK
OK
Normality
Rejected
Rejected
Rejected
Equal variance
OK
OK
OK
Model 1 Auto sales 58,13.639 1.100639
(Exports of manufactured goods)
10
Time-Series Decomposition
11
Time-Series Decomposition
F
Time-Series Decomposition Analysis cont.
12
Time-Series Decomposition
13
Time-Series ARIMA
14
Regression vs. Time-Series
15
Managerial Summary
Jan. 1997 . . . . . . . 51,412.70 Feb. 1997 .
. . . . . . 51,734.50 Mar. 1997 . . . . . . .
52,056.20 Apr. 1997 . . . . . . . 52,377.90
Write a Comment
User Comments (0)
About PowerShow.com