
Research Article


Numerical Analysis of the Impact of Science and Technology Investment on Regional Economic Growth in the Typical Agricultural Metropolis, Bozhou City, Anhui Province, China 

Kerong Zhang
and
Wuyi Liu



ABSTRACT

The Error Correction Model (ECM) modeling and a series of econometric tests were used to compute and estimate the impact and contribution of local science and technology investment on regional economic growth in Bozhou, China. Numerical analysis found that the correlation between regional economic growth, local science and technology investments was significant both in shortterm fluctuation and longrun equilibrium in Bozhou. The overall empirical result revealed that the impact or contribution of science and technology investment on regional economic growth was positive and it proved to be true that the variable lnSTINV_BZ (Science and Technology Investment in Bozhou) did Granger caused the variable lnGDP_BZ (Gross Domestic Product in Bozhou) in the ECM modeling. It was estimated that the elasticity coefficients of the contribution of local science and technology investment on regional economic growth were 0.175065 in the long terms and 0.421282 in the short terms. Then, the Q statistic was computed as 1.069091 in Bozhou. It was inferred that the regional growth of GDP in Bozhou was promoted by local science and technology investment at the estimated rates of 0.421282 (42.1282%) in the short terms and corrected or adjusted by 0.175065 (17.5065%) in the long terms, when the science and technology investment was increased by 1% accordingly. Furthermore, it was also found the promoting efficiency of local science and technology investment on the regional GDP in Bozhou was high according to the Q statistic.





Received: October 02, 2015;
Accepted: November 19, 2015;
Published: March 15, 2016


INTRODUCTION
Recent years see the rapid economic growth of different districts or regions in China (Xu, 2010; Tan, 2010; Fu and Gong, 2011; Fu et al., 2011; Aixiang, 2011; Hu et al., 2014; Yu and Ju, 2011; Zhong et al., 2011; Yu, 2012; Qiao and Zhu, 2012; Mao et al., 2013; Li and Wang, 2013; Chen et al., 2013; Yuan, 2013; Liu et al., 2014). However, the whole country of China is generally economically undeveloped, especially cities and counties in the midland districts with overall less developed industries and large agricultural populations, such as Fuyang, Bozhou and Suzhou in the Northern region of Anhui. These are typical agricultural districts of undeveloped and developing provinces in Central China. In particular, a series of supportive documents with 10 preferential policies in the governmental files had been issued by the provincial government to promote the economic development of three particular cities (i.e., Fuyang, Bozhou and Suzhou) and seven specific counties in Northern Anhui (i.e., Fengyang, Shouxian, Huoqiu, Huaiyuan, Wuhe, Guzhen and Suixi) and other counties during 20082014 (PGAP., 2008, 2010, 2014). Therefore, the local science and technology investment and its impact on the economic development of regional society and community are hot topics concerned by the people and the government and the utmost targeted achievements of the provincial government are to obtain the rapid development, expansion and leapfrog of the economic status and the transformation of science and technology in the Northern districts of Anhui among those special documents and policies (PGAP., 2008, 2010, 2014).
It is well known that science and technology is one of the most important influential factors and crucial linkers in the social production chain. It is agreed that the advanced science and technology will bring positive economic growth and the development of science and technology requires investment in science and technology. On the other hand, Anhui is one of the typical central agricultural districts in China and its regional investment of science and technology is a representative of Chinese agricultural areas. Therefore, research on this topic is particularly important for the government and its sectors to develop and make relevant regulatory policies and economic rules and regulations and/or in Central China. In general, investment, consumption and trade export are often referred as the three essential factors to drive economic developments in the highspeed development of modern economy. It is widely recognized that the differences in regional productivity are mainly accelerated or driven by the scientific and technological changes. Therefore, science and technology innovation is a crucial essential element of industrialization and informatization in China and other developing countries (Fan et al., 2004; Zhu, 2004; Meng et al., 2004; Su et al., 2006; Fu and Yang, 2009; FortuneDevlaminckx and Haunschmied, 2010; Fu and Gong, 2011; Fu et al., 2011; Zhang, 2013; Liu et al., 2014). At present, although there was a substantial concern on the impact of Chinese domestic investment on the developments of local regions and/or firms, most of the articles focused on Foreign Direct Investment (FDI) and there is only a few reports exploring the impact and/or policy development of science and technology investment in the undeveloped countries or areas (Liu and Buck, 2007; Li, 2009; Yang et al., 2009; Xu, 2010; Tan, 2010; Fu and Gong, 2011; Fu et al., 2011; Aixiang, 2011; Hu et al., 2014; Yu and Ju, 2011; Zhong et al., 2011; Yu, 2012; Mao et al., 2013; Li and Wang, 2013; Zhang, 2013; Yuan, 2013; Liu et al., 2014). Co (2000) reviewed the activities and conditions R and D, foreign direct investment and technology sourcing in USA during the 1980s. Tse (2001) analyzed the distribution of demand, market structure and investment of technology in theory. Tassey (2004) studied the policy issues for R and D investment in the knowledgebased economy of USA. Ane et al. (2007) analyzed the impact of technology assimilation on investment policy of R and D. Liu and Buck (2007) empirically investigated the impact of different channels for international technology spillover of Chinese hightech industries using panel data. Ko and OseiBryson (2008) reexamined the impact of information technology investment on hospital productivity by regression equations. Hritonenko (2008) reported the results of modeling the optimal investment in science and technology with nonlinear methods. Ratchford and Blanpied (2008) compared and analyzed different paths to the future for science and technology in China, India and USA. Olsen et al. (2008) analyzed the policies, paradigms and practices shaping US research and development and their evolution tracks. Li (2009) empirically analyzed China’s regional innovation capacity in transition. Yang et al. (2009) compared the R and D productivity of technologybased firms (NTBFs) located within and outside of science parks by measuring the elasticity of Taiwan R and D. Fu and Yang (2009) explored the crosscountry differences in patenting with a panel data of 21 OECD countries over the period of 19902002. Xu (2010) reported the regional differences of science and technology investment ability of large and medium industrial enterprises in China. Tan (2010) established an innovation model and empirically analyzed the clean technology R and D and innovation in China too. Fu and Gong (2011) and Fu et al. (2011) analyzed the roles of foreign technology and indigenous technical innovation in the emerging economies and/or developing countries like China. Motoyama et al. (2011) reported the national nanotechnology initiative of USA. Aixiang (2011) studied the relationship among energy consumption quality, education and R and D funds (science and technology) based on the grey relation theory of Jiangsu, China. Hu et al. (2014) compared the R and D efficiency and the national innovation system across 24 nations from 19982005. Gibson and Naquin (2011) explored the efficiency of investment in innovation and the global competitiveness of Portugal. Yu and Ju (2011) investigated the configuration status of science and technology resource in the coal industry of Heilongjiang, China. Zhong et al. (2011) evaluated the performance of regional R and D investments in China with DEA models. Yu (2012) analyzed the interaction of the local economic growth and the investment of science and technology in Fujian province based on the vector error correction model. Li and Wang (2013) explored the correlation between the local economic growth and the investment of science and technology in Shanxi, China. Arguedas (2013) examined whether it was socially desirable that fines for exceeding pollution standards in the polluting firms. Mao et al. (2013) studied the science and technology investment of Chinese agricultural research institutions in national agricultural research. Chen et al. (2013) investigated the effects of science, technology resources investment on the differentiated regional economic growths in China. Yuan (2013) reported the nonlinear relation between the government science, technology investment and economic growth in China. Zhang (2013) reviewed the function and role of the science and technology investment system from the practices of South Korea and Taiwan. Liu et al. (2014) checked the industrial characteristics affecting industry innovation activities in Chinese hightechnology industries from 19982008. Shah et al. (2014) analyzed the impact of deploying suitable preservation technology for an inventory system. In fact, the global GDP shares accounted for R and D in industrialized countries have been growing steadily over the past two decades.
According to the national policy for the development of science and technology of China (20062020), it was proposed that the overall goal was designed as the following specific objectives by the year 2020: The R and D expenditure of the whole society will hit the amount of more than 2.5% of GDP and the contribution of science and technology to China’s economy will account for more than 60% and the dependence on foreign science and technology will reduce to 30% or less. However, the shortage of R and D funds is common in the underdeveloped regions of China. Thus, it is worthy of studying concerning how much will the limited R and D funds contribute to the regional economic development. In the present study, we explored the impact of the science and technology investment on local economic productivity using econometric tests and annual datasets collected in Bozhou, China. The study was designed to determine whether there was a strong correlation between lnSTINV_BZ and lnGDP_BZ. It was also targeted to analyze and estimate the positive impact or contribution of local science and technology investment on regional economic growth in Bozhou, China. The overall findings suggested that there was a statistically positive correlation between the amount of local science and technology investment and regional GDP in Bozhou and analysis of the related variables found the potential promoting efficiency of local science and technology investment on the regional GDP was very significant in Bozhou. The present study provided new insights into the impact of local science and technology investment on regional economic growth in Chinese rural areas.
MATERIALS AND METHODS
Data acquisition and preprocessing: The time series data of science and technology investment and regional Gross Domestic Product (GDP) were all collected annually with references to "Anhui Province Statistical Yearbook" and "Bozhou City Statistical Yearbook" during the years 19972012 in the study (Table 1). According to the national economic expenditure accounting approach, the annual data of science and technology investment was transformed from the annual data of science and technology expenditure since these two types of data were expected to be equal in theory. Furthermore, the logarithms of the variables of annual data were analyzed in this study. After the acquisition of original data, we processed and transformed these annual data with Consumer Price Index (CPI) based on the basic price of 1990 in order to eliminate the possible heteroscedasticity from the annual data of the regional GDP and the science and technology investment. It should be noted that other impacting factors involved were regarded as minor factors and ignored in the subsequent regression modeling and correlation analyses. Therefore, there might be autocorrelations of variables in the modeling and correlation analyses of the regional GDP and the science and technology investment. The transformed logarithmetic data of Bozhou and the corresponding correlogram analyses were showed in Fig. 1 and 2 referred to as the abbreviation of BZ.
Econometric theory and model tests: The present study carried out unit root tests, cointegration tests and the Error Correction Model (ECM) to make a systematic computing and modeling of the correlations between local science and technology investments and regional economic growth magnitudes in the northern seven cities of Anhui, China (Gao, 2006).

Fig. 1: 
Logarithmetic data of the regional GDP (lnGDP_BZ) and the science and technology investment (lnSTINV_BZ) during 19972012 in Bozhou 
Table 1:  Annual data of the regional GDP, science and technology (S and T) investment and CPIs collected in Bozhou city during 19972012 

S and T: Science and technology, GDP: Groos domestic product, CPI: Consumer price index, Unit: Ten thousand yaun 
Actually, the shortterm fluctuation and longrun equilibrium correlation between the economic growth and the science and technology investment would be revealed by ECM modeling. But before executing the econometric computing of ECM equation, the variables should be tested with unit root tests of Augmented Dickey Fuller (ADF) and cointegration tests whether they were suitable to be applied in the modeling. The subsequent steps were carried out using the following unit root tests and cointegration tests and theoretical modeling with some brief descriptions here. Unit root tests of ADF: The data of stationary sequences appears in fluctuations around a mean value and its tendency is clear to a mean value and easy to be tested. If a variable is stated in the stationary sequence, it can be represented as I (0). Meanwhile, it can be denoted as I (1), if its firstorder differential variable is stated in stationary sequence. Similarly, it can be denoted as I (2), if its secondorder differential variable is stated in stationary sequence. In the unit root tests of ADF, the null hypothesis is that a unit root exists and the sequenced variable or time series data is nonstationary. Therefore, these types of tests are all called unit root tests.

Fig. 2(ab): 
The correlogram analyses of the variables, (a) lnGDP_BZ and (b) lnSTINV_BZ 
In the study, for any sequenced variable, the general formula of ADF unit root test is defined as Eq. 1:
where, the letter "a" is a constant, "t" is a variable of time trend and "p" means the number of lags. The null hypothesis (H_{0}) of the test is that r = 0, while the alternative hypothesis (H_{1}) is that r≠0. If the null hypothesis is accepted, then there exists a unit root within the variable to be tested, i.e., it is in nonstationary series. Otherwise, the alternative hypothesis should be accepted and there is no unit root within the variable tested, i.e., it is stationary. Cointegration tests: The cointegration test was first proposed by the economists Engle and Granger to test the hypothesis whether, there was the cointegration relationship between two sequenced variables. Therefore, the main idea of Engle and Granger can be expressed as the following. If two or more variables are in linear combination of stability, there should be a longterm stable relationship and/or correlation between these variables, i.e., cointegration, even though the variables are in nonstationary series (i.e., each variable has a unit root). The commonly method of the cointegration test are the EG twostep method and the Johansen method. In the present study, the EG twostep method of the cointegration test were used to test the longterm stable relationship (or correlation) between GDPs and the science and technology investments in the seven northern cities of Anhui, China. The first step of the EG twostep method is to establish the ordinary least squares regression equation using the target variables. The next step is to test if the sequenced residual of the regression model established above is stationary with the unit root tests. After these tests, if the sequenced residual data is stationary and smooth, there is to be a cointegration relationship between these variables in the initial unit root tests. Thus, we can analyze the longterm stable relationship and/or correlation between these variables and the subsequent issues with the ordinary least squares regression equation established.
Error correction modeling: If the firstorder or secondorder differential variables for both the endogenous and exogenous variables are stated in the stationary sequences and cointegrations, Eq. 2 can be deduced to estimate the longrun equilibrium or relationship between the variables:
The corresponding ECM modeling equation can be expressed as Eq. 3: where, the mark “Δ” stands for the first difference variable and the symbol "ecm" is the longterm regression equation residual. Estimation of the contributing efficiency of local science and technology investment on regional economic growth: In order to exactly estimate the contributions and statistics of local science and technology investments on regional economic growths in the seven Northern cities of Anhui, a novel variable is introduced and defined as the promoting or contributing efficiency, which means how much does the science and technology investment contribute to the increased shares of the regional economic gain (GDP). This assumptive variable can denoted as a statistic Q:
where, n is the reported years (n = 19972012), variable S_{i} is the annual share of the contribution of local science and technology investment in the enlargement of regional economic growth (i.e., the annual share of lnSTINV expressed in lnGDP), while the variable β is the regression coefficient in the corresponding ECM regression model. Actually, the variable β is redefined here as the elastic coefficient of the contribution of science and technology investment (lnSTINV) on regional economic growth (lnGDP). If Q<1, it represents the contribution or impact of the variable (lnSTINV) is negative and at a low efficiency, i.e., the increased share of GDP accounts for less than the amount of the variable itself. If Q>1, it indicates the contribution or impact of the variable (lnSTINV) is positive.
RESULTS AND DISCUSSION
Unit root tests of ADF: In this study, all the data (19972012) was dealt with the econometric software package Eviews 6.0 (Quantitative Micro Software Inc., USA). As it could be seen from the unit root test results (Table 2), the ADF test statistics showed that both the variables of the regional GDP and the science and technology investment (such as ΔlnGDP, ΔΔlnGDP, ΔlnSTINV and ΔΔlnSTINV) were analyzed and revealed in stationary sequences at the levels of the firstorder or secondorder differences, i.e., I (1) and I (2). Although, it was indicated that the variables themselves were nonstationary, their firstorder and secondorder differences were stationary and the subsequent model tests and analyses were suitable and applicable.
Cointegration tests: From the stationary analyses of the variables (Table 2) computed for Bozhou, it was obvious that the variables of the regional GDP and local science and technology investment were stationary with no unit roots at both the levels of the firstorder and secondorder differences, respectively. Actually, unit root tests (Table 2) and cointegration tests (Table 3) on the regression residual variables and the subject variables (i.e., lnGDP and lnSTINV) should be made before executing the ECM modeling and other econometric tests. Then, it was found out exactly the longrun equilibrium relationship between these two types of variables by the cointegration tests on the regression residual variable in the following least squares regression Eq. 4 in the EG twostep analysis:
Table 2:  Stationary tests of the regional GDP and the science and technology investment during 19972012 in Bozhou 

Variables containing lnGDP and lnSTINV showed the relevant estimates for the regional GDPs and the science and technology investment in Bozhou, China. *, **Significant values at the levels of 5 and 1% estimated by EViews 6.0, respectively. The lag order p was determined in accordance with the AIC rules 
Table 3:  Stationary test of the estimated regression residual variables 

Variable containing ε showed the estimation of residual variable in Bozhou (abbreviated as BZ). *, **Significant values at the levels of 5 and 1% estimated by EViews 6.0, respectively 
where, the tstatistics for the variable (lnSTINV_BZ) and the constant (C) were 4.326859 (very significant; probability = 0.0007) and 18.11976 (very significant; probability = 0), respectively. In addition, the regression coefficient for lnSTINV_BZ was revealed as 0.421282. Furthermore, the statistics R^{2}, (i.e., adjusted R^{2}), DW (i.e., DurbinWatson) statistic and F were estimated as 0.572149, 0.541589, 1.146637 and 18.72171 (probability of Fstatistic = 0.000696, very significant), respectively. It was suggested that there appeared little autocorrelation in the variables according to the DW statistic (DW = 1.146637) and the stationary test of all the estimated regression residual variables analyzed in Bozhou (Table 3).
In addition, the prephase analysis suggested the hypothesis that science and technology investment positively promotes and contributes regional economic growth in Bozhou, Anhui, which agreed with those reported cases of other undeveloped provinces or districts in China (Su et al., 2006; Xu, 2010; Yu, 2012; Yuan, 2013). Therefore, ECM modeling was implemented in the subsequent parts.
ECM modeling: The results of the ECM modeling and other econometric tests for the selected four cities were showed and developed in detail in the following parts. The ECM modeling of the contribution of local science and technology investment on regional economic growth in Bozhou was carried out with the previously established least squares regression equation as Eq. 5:
where, the tstatistics for the variable (lnSTINV_BZ) and the constant (C) were 4.326859 (very significant; probability = 0.0007) and 18.11976 (very significant; probability = 0) and the regression coefficient for lnSTINV_BZ was 0.421282. Moreover, the two variables, i.e., the regional GDP and the Science and Technology Investment (STI) should be regarded as cointegrated only if the variable ε_{t} of the regression equation was found to be stationary in the unit root tests. The error term ε_{t} could be expressed as Eq. 6:
It was found that there was a cointegration between the variables lnGDP_BZ and lnSTINV_BZ through the unit root tests of the error term ε_{t}. Subsequently, the ECM modeling equation was established according to the theorem of Engle and Granger with the variables in cointegration analyzed above. It should be noted that the variable ecm was computed according to the equation for the error term ε_{t}. As it was found that the residual variable had a strong trend of correlation in the firstorder difference, the appropriate lagging items were added into the established ECM equation. Therefore, the final corrected ECM regression equation was expressed as Eq. 7:
where, the tstatistics for the variable (lnSTINV_BZ) and the constant (C) were 0.141327 and 0.181771, while the regression coefficient and tstatistics for the residual variable ecm were revealed as 0.175065 and 0.911671, respectively. Moreover, the statistics R2, DW and F were were estimated as 0.345239, 0.236113, 2.271106 and 3.163653 (probability of Fstatistic = 0.078794, significant), respectively. The DW statistics was 2.266063, which suggested that there was no autocorrelation. Then it could be concluded that the equation was acceptable and there was no autocorrelations for the variables due to shorter sample interval and smaller values of DW tests in the ECM equation.
Estimated contributing elasticity and the related coefficients of local science and technology investment on regional economic growth in Bozhou: In the regression equation of ECM modeling, the differential coefficient could be regarded as the elasticity of the contribution of local science and technology investment on regional economic growth in Bozhou, Anhui, China. This coefficient was considered as the shortrun elasticity coefficient. In the ECM regression equation, the differential coefficient of the variable of lnSTINV was at the relatively low level (i.e., 0.421282) in Bozhou. Furthermore, in the corrected ECM regression equation, the coefficient of the variable ecm was regarded as the longrun elasticity coefficient. That ecm coefficient was computed as 0.175065 in Bozhou. Thus, it was showed that the elasticity of the contribution of science and technology investment on regional economic growth was 0.175065 in the long terms and 0.421282 in the short terms in Bozhou, Anhui, China. These elasticity coefficients of the variable ecm revealed the significant capacity of longterm equilibrium error controlling and driving the growth of GDP. Whenever, the science and technology investment was increased by 1%, the regional growth of GDP would be elevated by 0.421282 (42.1282%) in the short terms and corrected or adjusted by 0.175065 (17.5065%) in the long terms accordingly with the elasticity coefficients estimated in Bozhou. Actually, there were many researchers reported similar efficient effect or impact of science and technology investment and technical innovation (Liu and Buck, 2007; Li, 2009; Xu, 2010; Fu and Gong, 2011; Fu et al., 2011; Yu, 2012; Li and Wang, 2013; Mao et al., 2013; Liu et al., 2014). Liu and Buck (2007) empirically investigated the impact of different channels for international technology spillover on the innovation performance of Chinese hightech industries and found that learning by exporting and importing effectively promoted innovation in Chinese indigenous firms. Li (2009) made an empirical study of China’s regional innovation capacity in technical transition using a stochastic frontier model to estimate and explain the increasing disparity in innovation performance between Chinese regions. Their estimated results show that government support, the constitution of the R and D performers and the regional industryspecific innovation environment are significant determinants of innovation efficiency due to the large difference in the firm’s innovation performance across the regions. Xu (2010) analyzed the regional differences of science and technology investment ability of large and medium industrial enterprises in China and found the regional capacities of science and technology investments were positively correlated with the economic growths and development potentials of different provinces. Fu and Gong (2011) and Fu et al. (2011) reported that the role of foreign technology and indigenous innovation was crucial in the emerging economies and/or developing countries like China. Yu (2012) and Li and Wang (2013) analyzed the positive correlations of the local economic growths and the investments of science and technology in Fujian and Shanxi, China, respectively. Their analyses found that the financial investments and expenditures of science and technology could promote the provincial economic growths both in the short and in the long terms to some extent, but the effects of short term financial investments and expenditures of science and technology on the economic growths were relatively less than that of the long term. Specifically, among all the input elements, the investment of science and technology was the most significant impacting influential factor of economic growth and the personnel of science and technology was regarded as the second key influencer. Mao et al. (2013) studied the science and technology investment of Chinese agricultural research institutions and found there were serious shortages of financial investment in the national agricultural science and technology. However, Zhong et al. (2011) evaluated the performance of regional R and D investments in China and found the increasing returns to scale had not yet occurred in any provinces, while the constant returns to scale had prevailed in most provinces in the Western region, while decreasing returns to scale had prevailed in most provinces in the Eastern and central regions. Chen et al. (2013) reported that there were differentiated regional economic growths due to the different effects of science and technology resources investments. They found that the science and technology resources were coming into differentiation. There were different contributions of science and technology resources investments to the regional economic growths of different districts at the provincial level, while the decreased social technological progress and variations in science and technology resources investments were the main reasons for the declined efficiency of economic growths in some of the provincial districts. Mao et al. (2013) checked the commercialization of Chinese nanotechnology from the intertwined perspectives of academiaindustry relations, government support and policy, role of venture capital and international connections. Their analysis showed that China’s effort to commercialize nanotechnology has been much slower than the anticipated by the decreases of investments in science and technology resources, despite the tangible success in publishing, patenting and the creation of dedicated nanotechnology parks. Liu et al. (2014) explored how the characteristics of industry affect industry innovation activities in Chinese hightechnology industries considering the impact of foreign competition on innovation activities at industry level in a large emerging economy. They found that the competition intensity of foreign invested enterprises and domestic skill intensity affected the industry buy and make activities and foreign competition was positively associated with the intensity of buy activity (of science and technology resources), but negatively affected the intensity of make activity (of technical innovations). Further, it was showed that domestic skill intensity weakened the impact of foreign competitive pressure on technical innovation activities.
Table 4: 
Estimated statistics of contributing efficiency of science and technology investment on economic growth in Bozhou 

Estimated statistic of the contributing efficiency of local science and technology investment on regional economic growth in Bozhou: In order to estimate the exact role and contribution of science and technology investment on economic growth, a novel variable was introduced and defined as the statistic Q, which meant how much does local science and technology investment contribute to the increased shares of the regional economic gain (GDP). In the provided case of Bozhou, the statistic Q was computed as 1.069091. It is indicated the contribution of the variable (lnSTINV) was positive and the increased share of GDP accounted for more than the amount of the variable (lnSTINV) in Bozhou. The statistic Q revealed that the promoting or contributing efficiency of local science and technology investment on regional GDP in Bozhou was relatively high due to its advanced agricultural science and technology (Table 4). This result also agreed with those reported cases of other undeveloped provinces or districts in China (Su et al., 2006; Liu and Buck, 2007; Li, 2009; Xu, 2010; Fu and Gong, 2011; Fu et al., 2011; Yu, 2012; Yuan, 2013; Li and Wang, 2013; Liu et al., 2014).
CONCLUSION
In the present study, the ECM modeling and other econometric tests were established and used to analyze and estimate the correlation and impact or contribution of local science and technology investment on regional economic growth in Bozhou, China. Numerical analysis found that the correlation between regional economic growth and local science and technology investments was significant both in shortterm fluctuation and longrun equilibrium in Bozhou from the point of view of econometric tests. The empirical result revealed that the impact or contribution of science and technology investment on regional economic growth was positive and it proved to be true that the variable lnSTINV_BZ did Granger caused the variable lnGDP_BZ in ECM modeling. Next, it was estimated that the elasticity coefficients of the contribution of local science and technology investment on regional economic growth were 0.175065 in the long terms and 0.421282 in the short terms. Moreover, the Q statistic was computed as 1.069091 in Bozhou. It was inferred that the regional growth of GDP in Bozhou was promoted by local science and technology investment at the estimated rates of 0.421282 (42.1282%) in the short terms and corrected or adjusted by 0.175065 (17.5065%) in the long terms, when the science and technology investment was increased by 1% accordingly. Furthermore, it was also found the promoting efficiency of local science and technology investment on the regional GDP in Bozhou was high according to the Q statistic. The present empirical study indicated that both the current and potential impact and contribution of local science and technology investment on regional economic growth was revealed as very significant in Bozhou.
ACKNOWLEDGMENTS
The authors would like to thank the anonymous referees for constructive suggestions. This work is supported by National Statistical Scientific Foundation (No. 2013LY051), the Anhui Provincial Educational 2016 Annual Project of Outstanding Young Talents Fund in Colleges and Universities (No. gxyqZD2016184), Anhui Provincial Educational Commission’s Key Project (No. SK2014A069), Fuyang City Social Scientific Project (No. FSK2015032), and Fuyang Normal University Educational Study Project (No. 2013JYXM24).

REFERENCES 
1: Ane, B.K., A.M. Tarasyev and C. Watanabe, 2007. Impact of technology assimilation on investment policy: Dynamic optimization and econometric identification. J. Optim. Theory Appl., 134: 321338. CrossRef  Direct Link 
2: Arguedas, C., 2013. Pollution standards, technology investment and fines for noncompliance. J. Regul. Econ., 44: 156176. CrossRef  Direct Link 
3: Chen, X.L., H. Li and F. Fan, 2013. Regional difference of economic growth of science and technology resources investment effect analysis. Resour. Dev. Market, 29: 1315, (In Chinese). Direct Link 
4: Co, C.Y., 2000. R&D, foreign direct investment and technology sourcing? Rev. Ind. Organiz., 16: 385397. Direct Link 
5: Fan, B.N., J. Lei and J.M. Luo, 2004. Research on relationship between the expenditure for science & technology and the economic growth. Sci. Res. Manage., 25: 104107, (In Chinese). CrossRef  Direct Link 
6: Fu, X. and Q.G. Yang, 2009. Exploring the crosscountry gap in patenting: A stochastic frontier approach. Res. Policy, 38: 12031213. CrossRef  Direct Link 
7: Fu, X. and Y. Gong, 2011. Indigenous and foreign innovation efforts and drivers of technological upgrading: Evidence from China. World Dev., 39: 12131225. CrossRef  Direct Link 
8: Fu, X., C. Pietrobelli and L. Soete, 2011. The role of foreign technology and indigenous innovation in the emerging economies: Technological change and catchingup. World Dev., 39: 12041212. CrossRef  Direct Link 
9: FortuneDevlaminckx, E. and J.L. Haunschmied, 2010. Diversity of firm's life cycle adapted from the firm's technology investment decision. Cent. Eur. J. Operat. Res., 18: 477489. CrossRef  Direct Link 
10: Gao, T.M., 2006. Econometric Analysis and Modeling: Reviews Applications and Examples. Tsinghua University Press, Beijing, China.
11: Gibson, D.V. and H. Naquin, 2011. Investing in innovation to enable global competitiveness: The case of Portugal. Technol. Forecast. Soc. Change, 78: 12991309. CrossRef  Direct Link 
12: Hritonenko, N., 2008. Modeling of optimal investment in science and technology. Nonlinear Anal.: Hybrid Syst., 2: 220230. CrossRef  Direct Link 
13: Hu, J.L., C.H. Yang and C.P. Chen, 2014. R&D Efficiency and the national innovation system: An international comparison using the distance function approach. Bull. Econ. Res., 66: 5571. CrossRef  Direct Link 
14: Ko, M. and K.M. OseiBryson, 2008. Reexamining the impact of information technology investment on productivity using regression tree and Multivariate Adaptive Regression Splines (MARS). Inform. Technol. Manage., 9: 285299. CrossRef  Direct Link 
15: Li, X., 2009. China's regional innovation capacity in transition: An empirical approach. Res. Policy, 38: 338357. CrossRef  Direct Link 
16: Li, X.L. and Z.J. Wang, 2013. Shanxi science and technology investment and economic growth grey correlation analysis. Sci. Technol. Ind., 13: 7477. Direct Link 
17: Liu, X. and T. Buck, 2007. Innovation performance and channels for international technology spillovers: Evidence from Chinese hightech industries. Res. Policy, 36: 355366. CrossRef  Direct Link 
18: Liu, X., I.R. Hodgkinson and F.M. Chuang, 2014. Foreign competition, domestic knowledge base and innovation activities: Evidence from Chinese hightech industries. Res. Policy, 43: 414422. CrossRef  Direct Link 
19: Mao, S., Z. Cao, Y. Liu and J. Wu, 2013. Study on science and technology investment of Chinese agricultural research institutions: Discussion on national agricultural research institutions investment in science and technology. Issues Agric. Econ., 1: 4956. Direct Link 
20: Meng, X.Y., X.H. Meng and G.R. Cheng, 2004. A positive research on the impact that economic growth to technical input. Inform. Sci., 22: 10191021. CrossRef  Direct Link 
21: Motoyama, Y., R. Appelbaum and R. Parker, 2011. The national nanotechnology initiative: Federal support for science and technology, or hidden industrial policy? Technol. Soc., 33: 109118. CrossRef  Direct Link 
22: Olsen, K.L., N.M. Call, M.A. Summers and A.B. Carlson, 2008. The evolution of excellence: Policies, paradigms and practices shaping US research and development. Technol. Soc., 30: 309318. CrossRef  Direct Link 
23: Qiao, J. and J. Zhu, 2012. Government Input to ocean science and technology and marine economic growth: Empirical study based on panel data. Sci. Techol. Manage. Res., 32: 3840. Direct Link 
24: Ratchford, J.T. and W.A. Blanpied, 2008. Paths to the future for science and technology in China, India and the United States. Technol. Soc., 30: 211233. CrossRef  Direct Link 
25: Shah, N.H., D.B. Shah and D.G. Patel, 2014. Optimal preservation technology investment, retail price and ordering policies for deteriorating items under trended demand and two level trade credit financing. J. Math. Modell. Algorithms Oper. Res., (In Press). 10.1007/s1085201492530
26: Su, Z.F., R.D. Hu and C.J. Yi, 2006. On the relation between the S&T investment and the growth of GDP: Empirical analysis based on theory of cointegration and VAR model. Sci. Technol. Manage. Res., 9: 2629. CrossRef  Direct Link 
27: Tan, X., 2010. Clean technology R&D and innovation in emerging countriesExperience from China. Energy Policy, 38: 29162926. CrossRef  Direct Link 
28: Aixiang, T., 2011. Research on relationship between energy consumption quality and education, science and technology based on grey relation theory. Energy Procedia, 5: 17181721. CrossRef  Direct Link 
29: PGAP., 2008. Several policies and suggestions on accelerating the development of some cities and counties in the northern Anhui province and along the Huai River Basin brought up by the provincial party committee and government of Anhui Province. The People's Government of Anhui Province (PGAP), The Governmental File of Anhui Province, China, November, 2008.
30: PGAP., 2010. Several opinions and suggestions on further promoting the development of the districts in the northern Anhui Province brought up by the provincial party committee and government of Anhui Province. The People's Government of Anhui Province (PGAP), The Governmental File of Anhui Province, China, May, 2010.
31: PGAP., 2014. Several opinions and suggestions on speeding up the sound and rapid development of the districts in the northern Anhui Province brought up by the provincial party committee and government of Anhui Province, China. The People's Government of Anhui Province (PGAP), The Governmental File of Anhui Province, China, November, 2014.
32: Tassey, G., 2004. Policy issues for R&D investment in a knowledgebased economy. J. Technol. Transfer, 29: 153185. CrossRef  Direct Link 
33: Tse, C.Y., 2001. The distribution of demand, market structure and investment in technology. J. Econ., 73: 275297. CrossRef  Direct Link 
34: Yu, X., 2012. On the economic growth and the investment of the science and technology in FujianBased on the analysis of vector error correction model. J. Fujian Commercial Coll., 3: 7176, (In Chinese). Direct Link 
35: Xu, Y.P., 2010. Regional difference analysis of science and technology investment ability of large and medium industrial enterprises in China. Sci. Technol. Manage., 12: 912, (In Chinese). Direct Link 
36: Yang, C.H., K. Motohashi and J.R. Chen, 2009. Are new technologybased firms located on science parks really more innovative?: Evidence from Taiwan. Res. Policy, 38: 7785. CrossRef  Direct Link 
37: Yu, B.L. and X.F. Ju, 2011. An investigation on configuration status of science and technology resource in the coal industry of Heilongjiang province. Energy Procedia, 5: 21672171. CrossRef  Direct Link 
38: Yuan, P., 2013. Chinese government science and technology investment and economic growth nonlinear relation research. Special Zone Econ., 8: 1518, (In Chinese). Direct Link 
39: Zhang, M., 2013. [Science and technology investment system: Experiences from the practices of South Korea and Taiwan]. Sci. Res. Manage., 34: 9499, (In Chinese). Direct Link 
40: Zhong, W., W. Yuan, S.X. Li and Z. Huang, 2011. The performance evaluation of regional R&D investments in China: An application of DEA based on the first official China economic census data. Omega, 39: 447455. CrossRef  Direct Link 
41: Zhu, C.K., 2004. [Dynamic relationships between government R&D expenditure and economic growth in China]. Sci. Sci. Manage. S & T., 3: 2933, (In Chinese). CrossRef  Direct Link 



