Printer Friendly

Empirical analysis on the influencing factors of price fluctuation in Xinjiang unginned cotton trading from the weather perspective.

1. Introduction

Chinese region formed different main agricultural areas, the agricultural products transaction price is the focus of each area of concern in the field of agricultural production, agricultural products prices not only by the influence of various economic factors, changes in weather conditions influence on the agricultural product transaction price exists objectively (Nelson, 2014; Fernandez, 2015). But the existing research mainly focused on the economy the factors and the social environment influence on the prices of agricultural products, the price fluctuation caused by meteorological environment changes and not too much attention (Ferris, 2005). The meteorological environment has obvious regional characteristics, in order to explain the problem, this paper selects Xinjiang and its main crop seed as the research object of this paper.

With the 2014 target price reform pilot in Xinjiang since the implementation of cotton trading price fluctuations, increase cotton farmers' transaction risk. Found in the actual investigation, the target price reform pilot work carried out after the correlation between the purchase price and the cotton seed cotton ginning factory trading level to further enhance the quality of cotton and cotton is not only affected by the growth period (from emergence to the first mature boll opening, growth period, about 120 days) on meteorological conditions, in the cotton boll opening period (from the beginning to the end for receiving flowers opening boll period) between the meteorological environment will also affect the quality of cotton. If the influence of various factors ultimately reflected in the price changes, then the degree of meteorological environmental impact on the trading price of cotton is? Due to the vast area of Xinjiang, there is a huge difference between the north and South Xinjiang natural environment and human conditions, in the cotton trade links, have spawned a variety of intermediary organizations, and Seed cotton intermediary transactions whether there is conducive to seed cotton trading price stability? Whether the cotton trading price formation in South and North Xinjiang regional differences are obvious different? In order to study the above problems, the Xinjiang cotton price data using October 2014 to December, and combined with the detailed meteorological observations form data samples; first determine the related indexes influencing factors of cotton prices, with the meteorological environment (temperature, rainfall, wind level), the quantity of the trade, intermediary, Xinjiang, North and South cotton trading level; then through the construction of the influence degree of the quantile regression model and OLS regression model for quantitative analysis of various factors in the model; finally based on the analysis results of summing up cotton transaction price operation characteristics, and combined with the target price reform pilot opinions put forward countermeasures to improve the subsidy plan.

In the field of agricultural economy, the previous studies mainly focus on the impact of economic environment on agricultural production, often ignoring the natural environment and agricultural price fluctuations in the link between. Most of the literatures focus on the analysis of the trading price of cotton (Goreux, 2007), the interaction mechanism between the factors related to research and Chinese cotton market price fluctuations, the international cotton price changes China found leading cotton price change; research on the Xinjiang cotton market is mainly concentrated in the income, industrial competitiveness, comparative advantages, environmental advantages and technical conditions etc.; research on the cotton target price reform work, some scholars of the existing pilot target price reform problems and suggestions are given a systematic explanation, and to estimate the price of cotton from the cotton price target, cost rate of return angle simulation.

The foreign research focused on macro meteorological environmental impact on agricultural production (Goodwin, 2005; Meyer,2008), Nelson et al based on agricultural production model from the angle of the influence of the weather, a comparative analysis of the measurement model and the parameters of the existing research, introduces the result of climate change and potential determinants, and focuses on the analysis of the model parameters in the past in the study of interpretation ability (Plato, 2005; Olga,2009). Francisco & Blanco by the economic impact of climate simulation, empirical research on the global and European analysis, found that the overall results of existing research in spatial memory than the small area in significant differences, and these differences make the model explain ability.

To sum up, the research on the price fluctuation of bulk agricultural products is less affected by the weather. As the main producing district China production prices of agricultural products and other regional disparity is an objective existence, and regional characteristics of agricultural product price fluctuation is strong, in order to make the interpretation of the model parameters are stronger, we will study the area down into specific regions of Xinjiang, and its main crops of cotton as the starting point.

2. Data sources and model construction

2.1. Data sources

The observation data are mainly concentrated in the various prefectures in Xinjiang County, effective transaction data from Xinjiang 24 counties, 13 counties in southern and Northern 11 counties, obtained the transaction data is 1140, the basic information of each variable are shown in table 1. The time for sample data collection was: "October 3, 2014" to "December 14, 2014".

2.2. Model building

1. Variable settings

Temperature([temperature.sub.i]): In general, the cotton picking needs to continue for a long time, in this period of time, the weather outside environment change will affect the quality of cotton grade, when the temperature is too high, the seed cotton in cottonseed oil will lead to cotton, yellowing, reduce its quality level; therefore, the suitable temperature is conducive to cotton picking and smooth transaction for.

Rain([rain.sub.i]): Cotton fiber is very sensitive to water, when the rain and snow weather, if you are not able to remove moisture timely, can also cause cotton color yellow, the lower the quality, but in the cotton picking period, in the rain and snow weather situation, showing sporadic characteristics, even if there is rain, has little effect on the quality of cotton in grade sunny conditions. If there is persistent rain and snow weather, it will seriously affect the recovery process of the seed, causing a serious impact on the level of cotton. Observation rainfall weather value is only 130 in the overall sample, accounted for only 11.4%; and the total sample of 740 data concentrated in the southern region, so the rain and snow weather effects on seed cotton price need to combine the model results further illustrate. But we need to pay special attention to explain, because for rain, snow and other weather data sample ratio of only 11.4%, in this case, will affect the farmers to sell seed cotton ginning factories and processing progress, in reducing the supply situation, cotton ginning factory in order to meet the processing capacity, with the objective to raise prices will.

Wind power rating([wind.sub.i]): When the blowing sand and floating dust weather, the emergence of a large number of impurities in cotton fiber, the impurity removal requires a lot of procedures, but also cannot be guaranteed to be completely removed, when the impurity rate is high, the quality of cotton grade decreased; when the wind level is too high, there will be a variety of debris, increase the content of heterogeneous fiber cotton fiber, reduce the quality of cotton; through the actual investigation found that wind has three levels, the effect gradually highlights. Therefore, the wind is higher than the "3" situation is set to "1", the other is set to "0"".

Transaction intermediary([agency.sub.i]): Found in the actual investigation, some farmers by means of transport, their own conditions and other factors, often need the help of "intermediary" finished cotton trading, there are four main types of transactions subject to the actual observed: brokers, farmers + cooperatives, companies, processing factory. Seed sell process, in the name of the company, factory purchase intermediary will help farmers to complete the transaction process, and directly to the fields of seed acquisition. "Farmers to plant cotton sold" and "to the fields of acquisition between quantity and price in the transaction are significantly different. Therefore, this article only defines the behavior of farmers to sell the cotton ginning factory as a "non intermediary", define the behavior of the acquisition process for "intermediary exists". The definition of "producer" transaction completed by itself as a "non intermediary" transactions subject to other definitions for the behavior of the "intermediate seed acquisition".

Regional differences between the north and the south ([nj.sub.i]): Xinjiang cotton area is far higher than the northern, southern and Northern Xinjiang due to different natural conditions, social structure and national situation, the southern and Northern cotton trading prices appear bigger difference. Through the analysis of statistical data, the average price in the transaction is 6.15 yuan /kg, the average transaction price in the 5.61 yuan /kg, Xinjiang cotton trading price was significantly higher than that of Xinjiang, and this is related to the southern and Northern Xinjiang cotton quality difference, southern drought, special natural environment more conducive to improving the quality of cotton.

Cotton trading ([grade.sub.i]): Effect of high grade cotton to cotton trading price of the large cotton trading price formation was significantly higher than other grades of long staple cotton, low grade cotton effects on seed cotton price is uncertain. Through the cross analysis found that the same length of different grades of seed cotton price gap is smaller. In addition to cotton outside, the average transaction price of 2128, 2129, 3128, 3129 seed cotton was 6 yuan /kg, 5.79 yuan /kg, 5.96 yuan /kg, 5.71 yuan /kg, of which 2128 and 3128 grade cotton average transaction price gap is not big, the 2129 and 3129 seed cotton trading price is close to that in the case of different grades of cotton the same seed cotton, cotton length average transaction price is close to the cause of this phenomenon is the main reason for the acquisition process of cotton ginning factory "deeply demand" is a common phenomenon.

Cotton trading volume ([volume.sub.i] and [lnvolume.sub.i]): The target price mechanism, ignored specific transaction price subsidies for farmers to get the amount and individual farmers, all farmers to obtain subsidies standards, resulting in subsidy funds allocated to make the number of transactions in larger groups benefit more, the price of this part of the population is higher than the average market price of the price recovery period, a large number of transactions the utility will get subsidy funds trading groups was significantly higher than the number of transactions in small groups.

2. Model building

Based on the above analysis, the dependent variable is the seed cotton trading price situation, construction factors affecting seed cotton price analysis model for this transaction:

[price.sub.i] = [a.sub.0] + [a.sub.1] [lnvolume.sub.i] + [a.sub.2] [temperature.sub.i] + [a.sub.2][rain.sub.i] + [a.sub.4] [wind.sub.i] + [a.sub.5][agency.sub.i] + [a.sub.6][nj.sub.i] + [a.sub.7] [grade.sub.i] + [[epsilon].sub.t] (1)

In this paper, OLS regression analysis and quantile regression analysis are used in the study, and their coefficients are compared. The study found that OLS regression analysis can't be used to describe the distribution of the trading price of cotton. Quantile regression method from different points on the analysis of the problem, a more detailed description of the conditional distribution, the analysis of the results of the interpretation ability is stronger. In order to investigate the influencing factors of cotton trading price on different quantiles, the quantile regression model is established as follows:

[Q.sub.q] ([price.sub.i] |[X.sub.i]) = [X'.sub.i][[beta].sub.q] (2)

In formula (2), the independent variable [X.sub.i], [[beta].sub.q] is the coefficient vector; [Q.sub.q] ([price.sub.i] |[X.sub.i]) the conditional quantile corresponding to the points (0 < q < 1) in a given situation. The corresponding coefficient vector is achieved by minimizing the absolute deviation:


In the actual process of analysis, this paper adopts bootstrap algorithm technology intensive quantile regression coefficients were estimated, and the sampling frequency is set to 400, that is to say by constantly with confidence interval sampling and sample, thus to infer the coefficient.

3. Empirical result analysis

3.1. Quantile regression results

This paper uses the bootstrap method to the seed cotton trading price quantile regression. The coefficient of sample variables in each quantile show different characteristics, so this paper chooses 9 percentiles, respectively is 0.1, 0.2, 0.3, 0.4, 0.50, 0.6, 0.7, 0.8, 0.9, 0.90, for comparison, also lists the results of OLS, as shown in table 2.

3.2. Quantile regression results chart

In order to further explain the effect of various factors on cotton trading prices, coefficient changes will be quantile regression by Figure 1 lists in detail, and combined with the analysis in figure 1.

3.3. Quantile regression analysis

1. Cotton trading volume. With the increase of quantile, the number of transactions on the trading price of seed cotton influence showed the first decline after rising "type U" trend, the trade expansion is obvious in cotton trading price promotion function. In the 10 to 80 quantile, influence the number of transactions on transaction price is relatively stable, after 80 quantile the coefficient fluctuated, in the 90 sub sites near the coefficients show the trend of rapid rising and falling rapidly, at sites close to 100, the number of transactions on the trading of cotton seed cotton effect the price increase decreased rapidly. Through the actual investigation found that the main reason is: after picking cotton can be stored in cotton ginning factory, such as high transaction price in cotton trading settlement, farmers are very sensitive to changes in the price of cotton trading, with cotton trading prices rise, farmers will be considered in its own right price a lot of seed node transaction, which makes the farmers have certain selectivity to the seed cotton price, the transaction price is high cotton trading volume will be significantly increased; another reason is the high price of cotton at the site are basically the goods long staple cotton, strong rigid demand leads to the high price.

2. Meteorological environment


Temperature change. The quantile results can be seen, temperature on the positive effect of cotton trading prices significantly, and the degree of the influence with quantile increased, the trading price of cotton showed that with the increase of temperature on the degree of influence in reducing price volatility. In the cotton trade in the process, the effect of temperature on seed cotton price is relatively stable, and the coefficient between 0 to 0.1, and no negative, show in Xinjiang during the cotton picking temperature is suitable, basically not on cotton trading price formation negative effect.

Rainfall weather. Quantile regression results and empirical analysis of the deviation, generally, rainfall will cause damage to the Seed cotton level, thereby reducing the transaction price of cotton. The model results show that in the 80 sub sites before the rain will not form a negative impact on the trading price of cotton, but showing a positive effect; in the 10 sub sites before the positive effect is stronger; between 10 and 80 percentiles, the effect is relatively stable, shows a downward trend in 80 points; after the site, there is a strong volatility, and negative effects. The reason is more complex, mainly lies in: one is in the period of cotton picking, the amount of rainfall in Xinjiang is relatively small, even if the rain weather, the duration is short, the moisture absorption of cotton Co., in a short period of time can bring rain water evaporate, little effect on cotton quality grade; two period in cotton picking, sporadic rainfall on the air in the dust, sand and other granular materials and some fibers have a certain effect, has certain effects on reducing impurities around; three is rain weather, will delay the cotton picking and sell, restrained seed supply and processing capacity of cotton ginning factory every day fixed, when the influence of rainfall weather on its processing ability is formed, cotton ginning factory will ensure its processing capacity by adjusting the cotton trading price way.

Wind power rating. The higher the level of wind, sand and dust in the air, the fibers and other impurities will be more easily into the seed, is equal to the impurity content in raising seed cotton, the wind level is higher than three, the effect will gradually highlight, and the model results also show that in the 90 sub sites before the wind to the trading price of cotton with a persistent negative, the coefficient of floating down in the -0.2, show that Xinjiang windy weather is not conducive to the cotton trade, and in the 90 quantile, the negative impact tends to zero, and then a sharp decline, close to -0.2, the main reason is that the high price is the main cotton sample sites.

3. Whether there is intermediary. In the previous 70 quantile, influence degree of Seedcotton intermediary transaction price is relatively stable, through actual investigation will be ginning factory in cooperation with brokers, cooperatives, companies and other intermediary priority transactions, and the priority is to achieve to suspend trading of farmers. Began to volatility in the 90 sub sites in the vicinity of the regression coefficient, the reason is that the turnover of all the cotton processing factory as "cotton", resulting in high transaction sites the main influence on cotton trading price effect is very big.

4. North and South Xinjiang. With the increase of quantile, "South Xinjiang" has more significant impact on the trading price of the cotton. In the 80 sub sites before the influence of seed cotton price in Xinjiang is relatively stable, at 80 points after the site began to volatility, indicating high prices at the site of the area affected by the strong. From the descriptive statistical analysis of the data found in the trading price of cotton farmers facing in North and South Xinjiang had significant differences, and the Southern is higher than the north. By calculation, the average price in the transaction in South Xinjiang is 6.15 yuan /kg, the average transaction price in North Xinjiang is 5.61 yuan /kg, South Xinjiang cotton trading price was significantly higher than north Xinjiang, the quality difference between the Southern and Northern Xinjiang is the main reason, the special natural environment conducive to improve the quality of cotton grade in south Xinjiang, the price is high at the high quantilehas the strong correlation with regional advantages in Xinjiang.

5. Cotton trading level. At the low point of cotton trading level, the influence on cotton trading price is uncertain, which is mainly manifested in: before the 30 quantile, cotton trading level coefficient did not pass the significance test, before the 80 quantile, the regression coefficient fluctuate around o. Show the level two and level three cotton before 80quantile, the transaction price gap is not big, the phenomenon is caused by lowered the price of the acquisition of seed cotton ginning factory. With the level of cotton trade increased, cotton trading level positive effect on cotton trading prices began to rapidly increase, such as near the 90 quantile the impact of level of cotton seed on cotton trading transaction price was significantly increased, the reason is the high grade cotton (long-staple cotton) has more significant impact on the trading price of cotton.

6. The results of high points

The regression coefficients of all the variables in the quantile model appear high volatility after 85 points, the main reason is that in high sample site data is mainly for the Long-staple cotton. Long-staple cottonhas special grade, special cotton planting area (mainly in Awati) and higher subsidies (subsidy standards (production part) is 1.3 times the other grades of cotton), higher consumer demand and other factors, because the cotton trading price is mainly affected by the impact of consumer demand, the role of other factors is limited, and showed a fluctuation features often violent, and the impact of factors other than the south north Xinjiang, the point is more closer to the quantile 100, the coefficient is close to 0, the reason is that the demand for Long-staple cotton remain at the higher level; and all the long-staple cotton samples are from south Xinjiang, The southern border of North and south, so with the improved of point quantile for the variable "North and South Xinjiang", the effect degree of increase is very large.

4. Conclusion

From the above research we can see that the changes of the meteorological environment not only affect the formation of cotton growth, will also affect the formation of cotton prices, temperature, wind, rainfall and other meteorological environment should not only be taken into account in agricultural production, the formation of prices of agricultural products in the ring section also needs to be attention. Intermediary organizations will exist in the business process of any agricultural products are, in the cotton trade in the process of existence is particularly wide, intermediary effect on the trade of agricultural products prices and cannot be generalized, need from the social environment to analyze the necessity and rationality in the cotton trade in the process, by the large number of seed cotton trading and the distribution characteristics of cotton ginning factory, resulting in some cotton farmers need the help of intermediary organizations to complete the transaction (cotton seed handling, transportation etc.). Therefore, the existence of Xinjiang cotton trading process intermediary organization has objectivity.

Influence of regional differences on any agricultural products cannot be ignored, the agricultural production, processing and consumption are not in the same area, the space position changes on the price of agricultural products is stronger, in this analysis we can see that the southern and Northern Xinjiang cotton trading price in the obvious differences, regional differences zoom into the country after the more obvious. Therefore, agricultural subsidy policy cannot be generalized, according to different regional characteristics to develop measures and schemes of the. The number and quality of agricultural products trade has decisive effect on the prices of agricultural products, from the analysis of the results can be seen, the number of transactions affect the trading price of cotton can be clearly analyzed, but the effect of cotton trading level influence on cotton trading price is not expected strong, the main reason is that in the process of cotton trading the "deeply demand" phenomenon, not only is the seed cotton, for other agricultural products, this phenomenon is widespread in the transaction process. Therefore, agricultural products trading links also need supervision and regulation.

Recebido/Submission: 15/05/2016

Aceitacao/Acceptance: 29/08/2016


This work was supported by National Social Science Fund Project of China, Project number: 15CMZ042. This work also was supported by Xinjiang key research base of Humanities and Social Sciences Major Projects, Project number: XJEDU030113A01.


Ferris, J. N. (2005). Agricultural Prices and Commodity Market Analysis. Michigan State University Press, 20-35.

Fernandez, J., Blanco, M. (2015). Modeling the economic impacts of climate change on global and European agriculture: Review of economic structural approaches. Economics--The Open-Access, Open-Assessment E-Journal, Kiel Institute for the World Economy, 9, 1-53.

Goodwin, B. K., Schnepf, R. (2005). Modelling soybean prices in a changing policy environment. Applied Economics, 37, 253-263.

Goreux, L., Plastina, A. (2007). The New ICAC Cotton Price Forecasting Model. Cotton: Review of the World Situation, 15-20.

Gonzalez, M., Gonzalez, L. (2015). The co-creation as a strategy to address IT governance in an organization. RISTI--Revista Iberica de Sistemas e Tecnologias de Informacao, (14), 1-15.

Meyer, L., MacDonald, S. (2008). Cotton and Wool Outlook. Outlook Report No. CWS08a, U.S. Department of Agriculture, 2008, Economic Research Service, 45-50.

Nelson, G., Mensbrugghe, D. (2014). Agriculture and climate change in global scenarios: why don't the models agree. Agricultural Economics, 45(1), 85-101.

Olga, I., MacDonald, S. (2009). U.S. Cotton Prices and the World Cotton Market Forecasting and Structural Change. Economic research report, 42-45.

Plato, G., Chambers, W. (2005). How Does Structural Change in the Global Soybean Market Affect the U.S. Price? Economic Research Service, 12-17. http://www.ers.

Jingzhou Wei *, Weizhong Liu

* Jingzhou Wei,

School of Economics and Business, Xinjiang Agricultural University, China
Table 1--Basic data information

variable          classification             count   Percentage

transaction       2.7-4.15 yuan/kg           32      2.80%
price of cotton   4.15-5.3 yuan/kg           227     19.90%
                  5.3-5.96 yuan/kg           341     29.90%
                  5.96-6.7 yuan/kg           442     38.80%
                  6.7-9.34 yuan/kg           98      8.60%

Cotton            0-16777.58kg               761     66.80%
trading volume    16777.58kg-25609.65kg      104     9.10%
                  25609.65kg-35423.06kg      57      5.00%
                  35423.06kg-600000kg        218     19.10%

North and         South =1                   740     64.90%
South Xinjiang    North =0                   400     35.10%

Intermediary      Yes(Producer)=0            476     41.80%
                  No(Brokers, growers +      664     58.20%
                  cooperatives, companies,
                  processing plants)=1

Cotton            3128=1                     933     81.80%
trading level     3129=2                     55      4.80%
                  2128=3                     20      1.80%
                  2129=4                     91      8.00%
                  Long-staple cotton =5      41      3.60%
Temperature                                  1140    100%

Rainfall          Yes=1                      1010    88.6%
weather           NO=0                       130     11.4%

wind power        Greater than 3 level=1     313     27.50%
                  Less than 3 level=0        827     72.50%
Total                                        1140

variable          classification             mean value   S.D.

transaction       2.7-4.15 yuan/kg           5.96         1.08786
price of cotton   4.15-5.3 yuan/kg
                  5.3-5.96 yuan/kg
                  5.96-6.7 yuan/kg
                  6.7-9.34 yuan/kg

Cotton            0-16777.58kg               25609.65     47470.66
trading volume    16777.58kg-25609.65kg

North and         South =1                   0.65         0.477
South Xinjiang    North =0

Intermediary      Yes(Producer)=0            0.58         0.493
                  No(Brokers, growers +
                  cooperatives, companies,
                  processing plants)=1

Cotton            3128=1                     3.01         0.585
trading level     3129=2
                  Long-staple cotton =5
Temperature                                  12.81        6.58206

Rainfall          Yes=1                      0.114        0.31799
weather           NO=0

wind power        Greater than 3 level=1     0.2746       0.44649
                  Less than 3 level=0

Table 2--Cotton price quantile regression coefficients

2.  Meteorological environment

Quartiles                 LOG (VOLUME)    TEMPERATURE    RAIN

OLS         coefficient   0.12O912       0.054739       0.183379
            SE            0.015769       0.004259       0.094068
q=0.1       coefficient   0.05681 **     0.074992 ***   0.404495 ***
            SE            0.024746       0.00648        0.103236
q=0.2       coefficient   0.026469 *     0.061015 ***   0.308952 ***
            SE            0.014161       0.004008       0.091279
q=0.3       coefficient   0.011966       0.060707 ***   0.312196 ***
            SE            0.O12744       0.002858       0.081715
q=0.4       coefficient   0.028964 **    0.057113 ***   0.280425 ***
            SE            0.011827       0.003631       0.060914
q=0.5       coefficient   0.040185 ***   0.O5O331 ***   0.220739 ***
            SE            0.O11453       0.003558       0.059982
q=0.6       coefficient   0.039368 ***   0.045746 ***   0.217696 ***
            SE            0.008118       0.003941       0.054238
q=0.7       coefficient   0.046691 ***   0.037553 ***   0.202834 ***
            SE            0.01178        0.004446       0.064145
q=0.8       coefficient   0.093341 ***   0.04277 ***    0.181155 *
            SE            0.030174       0.006999       0.09525
q=0.9       coefficient   0.223057 ***   0.022772 *    -0.048428
            SE            0.037655       0.01221        0.185814

Quartiles                 WIND            NJ             AGENCY

OLS         coefficient   -0.305942       0.572888       0.29571
            SE             0.064899       0.062987       0.05699
q=0.1       coefficient   -0.214908 ***   0.402302 ***  -0.208914 ***
            SE             0.059638       0.080282       0.057271
q=0.2       coefficient   -0.191293 ***   0.306515 ***  -0.155622 ***
            SE             0.04082        0.063369       0.044388
q=0.3       coefficient   -0.22915 ***    0.271158 ***  -0.059459
            SE             0.046264       0.050944       0.045271
q=0.4       coefficient   -0.178569 ***   0.229625 ***  -0.04757
            SE             0.041242       0.059591       0.043138
q=0.5       coefficient   -0.217817 ***   0.174658 ***  -0.028537
            SE             0.034737       0.041291       0.035997
q=0.6       coefficient   -0.29455 ***    0.184386 ***  -0.009471
            SE             0.034629       0.042728       0.040758
q=0.7       coefficient   -0.286281 ***   0.243207 ***   0-049635
            SE             0.046681       0.048133       0.0581
q=0.8       coefficient   -0.190993 **    0.463689 ***   0.327679 **
            SE             0.076678       0.085979       0.134582
q=0.9       coefficient   -0.132091       1.013556 ***   1.100304 ***
            SE             0.13853        0.120641       0.18342

Quartiles                 GRADE          C          Pseudo Ra

OLS         coefficient   0.100008       3-557393   0.2807
            SE            0.026544       0.154941
q=0.1       coefficient   0.013201       3-514      0.2792
            SE            0.027364       0.29438
q=0.2       coefficient   -0.041294      4.298676   0.2221
            SE            0.035192       0.168611
q=0.3       coefficient   0.048777       4.469895   0.1937
            SE            0.036838       0.121685
q=0.4       coefficient   0.060303 ***   4.484092   0.1732
            SE            0.021156       0.106781
q=0.5       coefficient   0.034532 *     4.698237   0.1489
            SE            0.017921       0.110496
q=0.6       coefficient   0.045768 **    4-876557   0.1207
            SE            0.022453       0.101686
q=0.7       coefficient   0.033107       4-991757   0.0959
            SE            0.023596       0.12487
q=0.8       coefficient   0.013218       4.626266   0.0929
            SE            0.083917       0.327733
q=0.9       coefficient   0.099958       3-654523   0.3008
            SE            0.075786       0.239389

Note: *, **, ***, respectively, expressed in the 1%, 5%, 10% level
significantly; the estimated value through the bootstrap method to get
the 400 iteration
COPYRIGHT 2016 AISTI (Iberian Association for Information Systems and Technologies)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wei, Jingzhou; Liu, Weizhong
Publication:RISTI (Revista Iberica de Sistemas e Tecnologias de Informacao)
Date:Nov 1, 2016
Previous Article:Architecture and design methodology of on-chip debug module for multi-cores system.
Next Article:Research on network curriculum resources recommendation system based on MVC technology.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters