Modeling and Forecasting the Textile Price Index Using Nonparametric Path Design - Innovative Management in Information and Production - page 242

Tabl e 3 The forecasting with two models

Forecasting with Forecasting with

Date The real values parametric model path design

June 2012 105.075 103.8326 105.2272

July 2012 104.83 102.9030 104.6778

Tabl e 4 The forecasting with nonparametric path design model

The monthly Consumer Retail Producer Benchmark The textile

average crude price price price one-year price

Situation oil price index index index deposit rate index

1 100.00 102.00 100.00 97.00 3.00 104.6017

2 100.00 103.00 100.00 97.00 3.00 105.0583

3 100.00 102.00 100.00 97.00 2.75 104.6017

4 100.00 103.00 100.00 97.00 2.75 105.0583

In order to forecast the index in August 2012, we use the observations from
January 2008 to July 2012 to estimate the constants in July 2012, meaning t 0 is July
2012. ( 8 ) is the equation.

y t = 0 . 751864 y t − 1 − 0 . 410953 y t − 2 + 0 . 449259 z t 5 + 0 . 018674 z t 7 + g ( Z t , T ) (8)

T

Z t =( z t 2 , z t 6 , z t 8 ) , t , T = 1 , 2 ,..., 55 .

41619,
45.72563, 1.81382, and 24.75634, respectively, which means the linear part is
more weighty than the nonlinear part. In light of the previous result, we do a
situation simulation as a forecasting to the index value in August 2012 presented
in Table 4 . We use the estimation of the coefficients of July 2012 to estimate those
in August 2012.

In July 2012, the five parts on the right side of ( 8 ) are 79.00210, − 43 .

5 Conclusion

The principal goal of this research is to offer model to forecast the textile price
index in China's textile market. Therefore, monthly average crude oil price,
consumer price index, retail price index, producer price index, and benchmark
one-year deposit rate are introduced as relevant factors with textile price index
by nonparametric path design, and it is identified that consumer price index and
producer price index influence the dependent in linear pattern still by nonparametric
approach. Because of the lagged effect of the textile price index, the lagged is
employed as linear independent variables. Therefore, we construct a path design
model with four linear independent variables and four control variable, namely,
nonlinear independent variables. Empirical tests show that the nonparametric path