Spurious Regression Research Articles

TIME-VARYING COEFFICIENT TAYLOR RULE AND CHINESE MONETARY POLICY: EVIDENCE FROM THE TIME-VARYING COINTEGRATION

(ProQuest: ... denotes formulae omitted.)1.INTRODUCTIONThe Chinese economy, known as the miracle, has maintained steady and fast growth for three decades, and attracts much attention from scholars and practitioners. Unlike Eastern European countries, China undergoing a gradual market-oriented reform in which its economic structure and regime are changing slowly and smoothly. For this reason, it is of great importance to study whether existing economic theory can be applied to China. In macroeconomics, the proposed by Taylor (1993), is an important monetary policy that specifies how the US Federal Reserve should adjust its targeted interest rate to the inflation and output gaps. Studying the validity of the Taylor rule for China is theoretically and practically important.There is a vast literature on the Taylor rule. For example, Taylor (1999) studied the validity of the Taylor rule using historical data covering 1897 to 1914 and 1955 to 1997, finding that the deviation of the actual interest rate from the targeted interest rate could help reflect the effectiveness of monetary policy. Kim and Nelson (2006) proposed a time-varying parameter forward-looking monetary policy rule and provided an efficient estimation using the Kalman filter. Yuksel, Metin-Ozcan, and Hatipoglu (2013) introduced an interest rate pass-through specification of the monetary transmission process in a general Taylor model. Xie and Luo (2002) were the first to check the validity of the Taylor rule for China, finding that it was a good measure for Chinese monetary policy and provided a reference for Chinese monetary policy implementation. However, Bian (2006) showed that the Taylor rule was unstable in China by using the GMM and cointegration test approach. Zheng and Liu (2010) developed a regime-switching Taylor rule with a time-varying inflation target, and showed that the Chinese monetary policy rule could be significantly divided into passive and active regimes.As Osterholm (2005) pointed out, however, previous studies ignored the nonstationarity of the variables in the Taylor rule model, despite the enormous body of literature. On one hand, the variables might not be stationary processes. Given the variable's unit root behavior, the cointegration relationships among them become essential for regression modeling. In the absence of cointegation, the model would be a spurious regression. On the other hand, the traditional fixed coefficient cointegration approach assumes a constant long-run equilibrium relationship among the variables, thus failing to capture the dynamic structure of the equilibrium relationship. China's gradual reform process is leading to a smooth structural change in the relevant economic variable and market, such as stock markets (Huang et al., 2000) and interest rate. For example, China is gradually carrying out interest rate liberalization. First, inter-bank lending and bond markets were deregulated in 1996; then, the ceiling on deposit rates and the floor on lending rates of commercial banks were relaxed; then, the Shanghai Interbank Offered Rate (Shibor) became China's benchmark rate in 2007; and then, the lending interest rate was completely decontrolled in 2013. Accordingly, this paper considers the smooth and gradual structural change occurring in China rather than abrupt structural breaks.This paper develops a time-varying coefficient Taylor rule to study China's monetary policy. Given the nonstationarity of the model's variables, the paper employs the time-varying cointegration approach of Park and Hahn (1999). To the best of our knowledge, this paper is the first to examine the Taylor rule through the time-varying cointegration approach. Kim and Nelson (2006) also considered the time-varying coefficient Taylor rule, but failed to consider the nonstationarity of the variables. By using several unit root tests, we found that the interest rate, the inflation gap and the output gap are indeed nonstationary. …

On spurious regressions with partial unit root processes

This paper considers spurious regressions with partial unit root and near partial unit root processes. Three models are investigated via simulations to study the rejection probability of the associated t-statistics in detecting the spurious correlation. The results reveal that the spurious regression is a not-so-spurious problem, which cannot be eliminated by adding AR(1) errors in re-estimation. Moreover, the advocated balanced regression approach is found effective in detecting the non-existent relationship.

Wavelet-cointegration prediction of irrigation water in the irrigation district

Affected by various influencing factors, many time series analysis methods are used to predict irrigation water. However, it is often assumed to be stationary for the hydrologic time series which will result in an insignificant “spurious regression”. Actually, irrigation water is largely determined by rainfall and crop water requirement in the irrigation district. With the wavelet analysis method, the varying fluctuation characteristics of rainfall, irrigation water, and crop water requirement with multi-temporal scales are exhibited. The cointegration equilibrium relationship amongst their original time series and decomposed time series are also revealed by the cointegration theory. Combining the wavelet analysis method with the cointegration theory, the wavelet-cointegration prediction model of irrigation water is proposed. The results show that the model has the available prediction accuracy, and the relative errors of all predicted years are less than 5%, except 2004 and 2012.

Evidence of spurious regression driven by heavy-tailed observations with structural changes

ABSTRACTThis article extends the theoretical analysis of spurious relationship and considers the situation where the deterministic components of the processes generating individual series are independent heavy-tailed with structural changes. It shows when those sequences are used in ordinary least-square regression, the convenient t-statistic procedures wrongly indicate that (i) the spurious significance is established when regressing mean-stationary and trend-stationary series with structural changes, (ii) the spurious relationship occurs under broken mean-stationary and difference-stationary sequences, and (iii) the extent of spurious regression becomes stronger between difference-stationary and trend-stationary series in the presence of breaks. The spurious phenomenon is present regardless of the sample size and structural breaks taking place at the same points or not. Simulation experiments confirm our asymptotic results and reveal that the spurious effects are not only sensitive to the relative location of structural changes with the sample, but seriously depend on the stable indexes.

Spurious regression due to neglected of non-stationary volatility

This paper examines the effects of permanent changes in the variance of the errors on routine applications of standard t-ratio test in regression models. It is shown the asymptotic distribution of t-ratio test is not invariant to non-stationary in variance, and the phenomenon of spurious regression will occur independently of the structure assumed for these time series. The intuition behind this is that the non-stationary volatility can increase persistency in the level of regression errors, which then leads to spurious correlation. Monte Carlo experiment evidence indicates that, in contrast to the broken level/trend case, the presence of spurious relationship critically depends on the location and magnitude of changes, regardless of the sample size. Finally, some real data sets from the Shanghai stock database are reported for illustration.

ESTAMAING PRICE AND EXPENDITURE ELASTICITIES FOR MAJOR FOODS IN EGYPT DURING THE PERIOD (1980-2014) USING THE LINEAR ALMOST IDEAL DEMAND SYSTEM

The objective of this research is to estimate price (direct and cross) and expenditure elasticities for major food commodities in Egypt. The food commodities are divided into six sub-models depending on the homogeneity within each group of commodities and the data availability quantities consumed and the corresponding retail prices for each individual item. Some commodities are aggregated in groups such as other beans, other vegetables, fruits(1), fruits(2) and oils. The Linear Almost Ideal Demand System (LAIDS) model is applied in estimating a system of demand equations for each group of food commodities. To avoid the invalid inference and spurious regression problems that may be created by non-stationary data series, the Fully Modified Least Squares (FMLS) estimator is utilized. The demand parameters satisfy the Engel aggregation, cournot aggregation, homogeneity and symmetry conditions. The results indicate that expenditure elasticities of the majority of food commodities/groups are less than the unity except fot the fruits2 (1.110), indicating necessity effects. This can be also interpreted as following the increasing of consumptions of these commodities is strongly connected with increasing of all income levels. The results also show that the own price elasticities for food commodities/groups are inelastic. For fruits 2, its inelastic own price elasticity still indicates that it tends to be very sensitive to price changes. The relative high cross price elasticities in all sub-models illustrate the strong substitute or complementary effects of the price change of one commodity one quantities consumed from other commodities in the same sub-models. Therefore, structural implications from the estimated elasticities are important. The estimated own and cross price and expenditure elasticities must be analyzed during the economic reform for better understanding all economic changes affecting prices as well as consumption and expenditures.

ELASTICITY OF SUBSTITUTION AND TECHNICAL PROGRESS: IS THERE A MISSPECIFICATION PROBLEM?

In previous work, we estimated a dynamic model of the Italian economy, showing that its weakness in the past two decades is mainly due to the slowdown in total factor productivity growth. In those models, two parameters play a key role: technological progress and the elasticity of substitution. Recent estimates of those parameters are affected, in our opinion, by a specification problem: technological parameters are inherently long-run but their estimates are based on short-run data. Looking deeply into the estimation procedure, we show that the misspecification issue present in the estimates gives rise to a spurious regression bias (high R2, low DW), because the standard approach does not incorporate frictions and rigidities. Our modeling strategy takes account of them. Although we cannot in general say that our framework gets rid of the serial correlation problem, the statistics for our model do show that residuals are not serially correlated.

A simple solution for spurious regressions

ABSTRACTThe literature on spurious regressions has found that the t-statistic for testing the null of no relationship between two independent variables diverges asymptotically under a wide variety of non stationary data-generating processes for the dependent and explanatory variables. This paper introduces a simple method which guarantees convergence of this t-statistic to a pivotal limit distribution, thus allowing asymptotic inference. This method can be used to distinguish a genuine relationship from a spurious one among integrated processes. We apply the proposed procedure to several pairs of apparently independent integrated variables, and find that our procedure does not find (spurious) significant relationships.

External Debts and Economic Growth in Nigeria: An Empirical Study Using Autoregressive Distributed Lag Model

This paper models some economic theories which explained the casual relationship between accumulated funds/loans from external sources (external debts) and economic growth with a more focus on Nigerian economy, as it is a usually trend for debts of third world and developing countries debts profile to be high, and this case is also the same with Nigeria as the country debts profile is on the increase once more after the debt cancellation of 2005 by the Paris Club of creditors. Also, the addition to empirical studies on this topic by this study is the use of Autoregressive Distributed Lag (ADL) model to capture the effect of externals debts on viability and growth of Nigerian economy from 1984-2012. The preliminary and normality tests show that all variables used in the study are positively skewed but are not normally distributes. For more diagnostic analysis of the dataset, econometric tests of stationarity (Unit Root Test) and co-integration tests were employed to avoid a spurious regression. The results of these tests indicated that all the variables exhibit Stationarity at first differenced, and also the Johansen Co-integration test shows the existence of long-run relationship between the variables. The result from the ADL model confirms the existence of a dual behaviour and effect of External Debts on Nigeria Economy. The lag 1 of external debts i.e Debts from previous, has positive effect while external debts of present year posed a negative effect on the performance of the economy. The recommendations of this study are that Nigeria government should ensure that debts incurred are channel towards productive uses and debt management office of the government should strengthen its plans and foster appropriate use of loans in the critical area the need for is identified.

Dividend persistence and dividend behaviour

AbstractThis article demonstrates how a spurious regression problem caused by dividend persistence is compounded by a spurious correlation problem when the dependent and independent variables in dividend behaviour regressions are ratios composed of common component variables. This article utilises a simulation procedure to take account of these problems, with the findings implying that extreme care should be taken when using ratios as predictor or explanatory variables in time series regression. This article introduces a reformulated Lintner first difference dividend behaviour model that is not subject to spurious regression in which past prices predict subsequent changes in dividends.

Prediction of primary energy demand in China based on AGAEDE optimal model

In this article, we present an application of Adaptive Genetic Algorithm Energy Demand Estimation (AGAEDE) optimal model to improve the efficiency of energy demand prediction. The coefficients of the two forms of the model (both linear and quadratic) are optimized by AGA using factors, such as GDP, population, urbanization rate, and R&D inputs together with energy consumption structure, that affect demand. Since the spurious regression phenomenon occurs for a wide range of time series analysis in econometrics, we also discuss this problem for the current artificial intelligence model. The simulation results show that the proposed model is more accurate and reliable compared with other existing methods and the China’s energy demand will be 5.23 billion TCE in 2020 according to the average results of the AGAEDE optimal model. Further discussion illustrates that there will be great pressure for China to fulfill the planned goal of controlling energy demand set in the National Energy Demand Project (2014–2020).

The application of production functions to Iranian higher education system

This article aims to empirically examine the application of production functions for evaluating remarkable developments in Iranian higher education system during the past three decades, using data from the Statistics Center of Iran and Institute for Research and Planning in Higher Education for the 1988-2014 period. A variety of production function forms have been used to evaluate the main determinants of Iran's higher education performance, and cointegration analyses have been conducted to rectify the problem of spurious regressions and to evaluate long-run and short-run relationships among the variables used. We found that: (1) estimated Iranian higher education production functions conform the polynomial cubic as well as Cobb-Douglas forms; (2) academic staff and number of institutions are the main determinants of higher education outputs in both long-run and short-run explaining more than 98% of the variations in the dependent variable, with the human capital factor playing a dominant role; (3) Iranian higher education system performs at the situation of increasing return to scale exhibiting some degree of inefficiency; and (4) the estimates of Cobb-Douglas forms provide evidence on the existence of substitutability among various academic staff.

A comment on ‘resolving spurious regressions and serially correlated errors’

In order to diminish size distortions of the t test in a time series linear specification, Agiakloglou (Agiakloglou in Empir Econ, 45(3):1361–1366, 2013) proposed to (1) include the first lag of the dependent variable as a regressor or (2) estimate it using the first differences of the variables. He provided finite-sample evidence to support his proposal. In this paper, we extend the Monte Carlo experiment to different data-generating processes and calculate the asymptotic behavior of the modified specifications. We show that including the lag of the dependent variable as a regressor reduces size distortions when the variables are driftless unit roots, but this approach does not hold under the presence of long memory, nonlinearities, or structural breaks.

Determinants of Private Investment in Tanzania: 1970-2015

This study provides an empirical analysis of the macroeconomic factors that affect private investment decisions in Tanzania. Theory and empirical literature is reviewed in an effort to identify a private investment function for the period 1970 to 2015. The results suggest that private investment is determined by aggregate demand as measured by output. In addition, there is evidence that the crowding in effect of public investment and credit flow to the private sector has positive impact on capital accumulation, while external debt and inflation have had negative effects. Since time series data can bring spurious regression results, we avoid this by testing for stationarity and co-integration prior to estimation of the model. The results of the ECM model suggest that private investment is co-integrated with the suggested variables. In terms of policy, the results suggest the importance of inflation control, development of credit markets and public investment as economic policy instruments for private investment growth in Tanzania.

An Empirical Analysis of the Impact of Unemployment on Economic Growth in Zimbabwe

This study examined the impact of unemployment on economic growth in Zimbabwe for the period 1982 to 2013. In this study we use the ordinary least squares (OLS) to estimate the regression equation. The first step is to make the data stationary to avoid spurious regression. Based on the stationarity test results, cointegration tests were conducted to test for the existence of a long run relationship between economic growth and its determinants. The Error Correction Model (ECM) was also employed to establish the short run dynamics and speed of adjustment to the long run. The estimation results for both the long run and short run models revealed that unemployment has a significant influence on productivity growth. Also shown from the results is the negative relationship between unemployment and economic growth.

Impact of Microfinancein Promoting Financial Inclusion in Nigeria

<p><em>Despite the Central Bank of Nigeria’s (CBN) initiatives to encourage banks to extend their services and facilities to rural areas, a high percentage of the rural dwellers still remain unbanked and as such, the initiatives appear not to promote financial inclusion services among the poor in Nigeria. As a result, small entrepreneurs often lack enabling financial environment to grow. The study undertakes an empirical analysis of the impact of microfinance in promoting financial inclusion in Nigeria between 1990 and 2014 using OLS regression method. Unit root test was conducted on the variables to examine their level of stationary to avoid spurious regression results. The findings showed that minimum deposit amount have a positive and significant relationship with saving. It was observed that access to microfinance minimum deposit amount has significant effect on savings account opened by rural dwellers. Microfinance interest rate was however found to have a negative and insignificant relationship with the rural dwellers loans and advances.</em><em> </em><em>Recommendations were made among which are that Government should facilitate microfinance branches close to the rural area, products and services accessible to a large segment of the potentially productive Nigeria population, who are currently not being served by the formal financial sector.</em></p>

Detecting the socioeconomic driving forces of the fire catastrophe in NW Spain

Wildfires cause devastating environmental, social and economic effects in different regions of the world. The aim of this study was to analyze the long-run relationship between the number of ignition events and socioeconomic variables using time series data. We focus on Galicia, a region in the northwest part of the Iberian Peninsula and with one of the highest fire density and largest burned areas in Europe. Since the late 1980s, the number of forest fires has increased in Galicia and caused extensive damage to the environment, property and human life. The analysis is based on cointegration tests between variables. In order to avoid the problems related to spurious regression, the ARDL bounds testing approach was applied. The statistical evidence allows us to conclude that in the long term, the price of eucalyptus timber, the population in the primary sector and the intensity of the elections are relevant factors in explaining the start of forest fires. These three variables are found to increase the propensity of the population to start a fire that cause devastating environmental, social and economic effects.

Monitoring Stationarity and Cointegration

We propose a monitoring procedure to detect a structural change from stationary to integrated behavior. When the procedure is applied to the errors of a relationship between integrated series it thus monitors a structural change from a cointegrating relationship to a spurious regression. The cointegration monitoring procedure is based on residuals from modified least squares estimation, using either Fully Modified, Dynamic or Integrated Modified OLS. The procedure is inspired by Chu et al. (1996) in that it is based on parameter estimation only on a pre-break ``calibration'' period rather than being based on sequential estimation over the full sample. We investigate the asymptotic behavior of the procedures under the null, for (fixed and local) alternatives and in case of parameter changes. We also study the finite sample performance via simulations. An application to credit default swap spreads illustrates the potential usefulness of the procedure.

New bootstrap inference for spurious regression problems

Phillips [11] provides asymptotic theory for regressions that relate nonstationary time series including those integrated of order 1, . A practical implication of the literature on spurious regression is that one cannot trust the usual confidence intervals (CIs). In the absence of prior knowledge that two series are cointegrated, it is therefore recommended that we abandon the specification in levels and work with differenced or detrended series. For situations when the specification in levels is sacrosanct we propose new CIs based on the Maximum Entropy bootstrap explained in Vinod and López-de-Lacalle (Maximum entropy bootstrap for time series: The meboot R package, J. Statist. Softw. 29 (2009), pp. 1–19). An extensive Monte Carlo simulation shows that our proposal can provide more reliable conservative CIs than traditional and block bootstrap intervals.

P-083 – Recueil d'urine chez moins de 1 mois par technique de stimulation

We examine whether military regimes harm stock market performance by investigating stock returns in ten emerging markets under military and civilian rule. We find no evidence of military regimes having a significantly negative impact on stock returns. In the case of Thailand and Pakistan, we find a significant positive military return premium. These returns cannot be explained by economic cycles, stock market cycles, or returns volatility. Our findings are robust to worldwide stock market movements, tests for spurious regression bias and randomization-bootstrap tests. Our results contradict the common view that military rule has a negative impact on stock market performance.