Appreciation Pressure Of Rmb Exchange Rate Economics Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Nowadays, more and more foreign countries have been turning the heat on appreciation of RMB exchange rate. The stress made by other countries results from China's large amount of trade surplus and vast influxes of capital brought by foreign investment. Another important reason is that foreign countries, such as America and Japan, try to relieve the pressure of domestic economic depression by forcing the appreciation of RMB. Consequently, the study of RMB exchange rate has great significance on today's economy. This paper proposes to make a prediction of the moving trends of the RMB exchange rate by using the method of Markov regime switching model. Firstly, this paper tries to use five influencing factors which are CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price to study the trends of RMB exchange rate. The monthly data of the RMB exchange rate middle price against USD, CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price from Jan 1994 to Dec 2010 are adopted. However, after using regression analysis, correlation analysis and some other tests by stata, it can be found that using these influencing factors to study the trends of RMB exchange rate is not appropriate. This result may be due to the macroeconomic coercive control of Chinese government. Then, this paper uses the data of RMB exchange rate itself to predict the trends by using matlab. The outcomes show that there were two regimes during the period from Jan 1994 to Dec 2010. The first regime is relatively stable with duration of 56.05 months while the second regime has a higher negative growth rate and a higher volatility with duration of 23.99 months. The results indicate that the moving trends forecasted by Markov regime switching model are quite matched with the reality. In the year 2011, RMB exchange rate may continue to decrease.

Keywords: Markov regime switching model, RMB exchange rate.

INTRODUCTION

Nowadays, Markov regime switching model has been used successfully for modelling financial time series (Liu and Zhang, 2010). Markov regime switching model aims to catch the discrete shifts in the behaviour of financial variables. In the process of the capture, the parameters of the underlying data-generating process are allowed to take distinct values in different time periods. In short, a framework is established with two states to describe two different prediction regimes. The fundamental principle behind the application of these models comes from the fact that the characteristic of the dynamic changes in RMB exchange rate may be the regime shifts. This indicates that the RMB exchange rate is permitted to be dependent on the "state of the market".

For the purpose of learning the dynamics of the regime switching of the moving trends, the RMB exchange rate is firstly modelled in regime switching within a univariate Markov-Switching model. It is emphasized that one of the most significant characteristics of the Markov regime switching model is to estimate the probabilities of each regime at one time point. The econometric methods of estimating the parameters in the Markov regime switching model have been extended from the previous studies. Besides, the researchers also confirmed how Markov regime switching model which is better than the existing single regime models can characterize time series behaviour of some variables.

The notion of Markov regime switching firstly dates back to Henderson and Richard (1958) who write the "Microeconomic Theory: A Mathematical Approach". Markov regime switching model is applied by Hamilton (1989) to the study of the United States business cycles and periodic changes of regimes from the positive growth rate to the negative growth rates in post-war US real GNP. Hamilton (1989) also developed Markov regime switching model to the case of auto correlated dependent data. The problem of whether the Markov regime switching model can forecast exchange rate or not is firstly put forward by Engel (1994). In that paper, Engel (1994) made use of a time series approach with two regimes and applied the model to three currencies. Moreover, Engel (1994) also pointed out that the direction of the change of the exchange rate could be predicted quite well by the forecast of the Markov regime switching model. Later, a Markov switching model for daily exchange rate data was used in Marsh's (2000) study in which the only one fundamental is the interest rate differentials. After finishing the study, Marsh (2000) summarized that compared with a pure time series Markov switching model, superior forecasts for the exchange rate cannot be provided through the method he used in this study. Frommel, MacDonald and Menkhoff (2005) proved in their studies that there were important regime changes in a monetary exchange rate model for the three most important currency markets, namely the bilateral USD values of the D-Mark, Yen and Pound. They came to the conclusion that there seems to be two highly persistent regimes that can be identified for the three exchange rates under investigation.

In this paper, the aim is to predict the moving trends of RMB exchange rate through the method of Markov regime switching model. We firstly try to use potential influencing factors to analyse the moving trends of RMB exchange rate. However, after using regression analysis, correlation analysis and residual test, we find that it is impossible to use the influencing factors when we predict the moving trends of RMB exchange rate. This is mainly because that RMB exchange rate is established and controlled by Chinese government instead of the market. As a result, we use RMB exchange rate itself to make a prediction of the moving trends. In the process of forecasting, two sorts of software which are Stata and Matlab help us to find out the final results of the prediction. By analysing the data of RMB exchange rate middle price against USD from Jan 1994 to Dec 2010, we make a conclusion that RMB exchange rate against USD is appreciated in these years. Besides, there are two downward regimes in the moving trends. Compared to the first regime, the RMB exchange rate in the second regime appreciates faster. The first downward regime holds the duration of 56.05 months while the second downward regime holds the duration of 23.99 months.

The paper is structured as follows: Section 2 presents some background information of exchange rate. Section 3 shows the data description and preliminary statistics. Section 4 describes the research methodology. Section 5 presents and analyses the empirical results. Finally, the conclusion is showed in Section 6.

APPRECIATION PRESSURE OF RMB EXCHANGE RATE

During Asian financial crisis in1997, Chinese government won the world's respect because of the persistence of keeping the RMB exchange rate steady. However, things have changed with the passage of time. Nowadays, more and more foreign countries have been turning the heat on appreciation of RMB exchange rate. The stress made by other countries results from China's large amount of trade surplus and vast influxes of capital brought by foreign investment. Another important reason is that foreign countries, such as America and Japan, try to relieve the pressure of domestic economic depression by forcing the appreciation of RMB. It is commonly recognized that the changes of RMB exchange rate have great influences on China's even the world's economy. As a result, people become to pay more attention to the trends of RMB exchange rate than before.

In this section, we will firstly introduce the influences of appreciation of RMB on China's economy, which includes two small parts, namely advantages of appreciation of RMB exchange rate and disadvantages of RMB exchange rate. Then, we will present the measurements which can be taken by Chinese government to deal with the pressure of appreciation of RMB exchange rate.

Influences of Appreciation of RMB on China's Economy

Advantages of Appreciation of RMB Exchange Rate

Changes of the RMB exchange rate will influence China's economy on various aspects. The advantages of appreciation of RMB are mainly revealed in the following three aspects.

Firstly, appreciation of RMB will excite the increase of import. When RMB exchange rate is appreciated, the prices of overseas consumer goods and raw materials are cheaper than before. This situation will be beneficial for reducing the costs of import.

Secondly, appreciation of RMB will improve the environment of attracting foreign investments. If RMB exchange rate is appreciated, the profits of overseas-funded enterprises which have invested in China before will increase. This growth of profits may heighten the confidence of investors, which will in return accelerate further investments or reinvestments. Besides, appreciation of RMB will also attract large amount of foreign investments to come into China's capital market and this behaviour will result in an increasing number of indirect investments.

Thirdly, appreciation of RMB will alleviate the burden of external debt. After RMB exchange rate is appreciated, the amount of outstanding external debt which includes principals and interests will comes down correspondingly. Therefore, the pressure of external debt can be reduced.

Disadvantages of Appreciation of RMB

Because of China's substantial trade surplus and vast influxes of foreign capital, the demand of the RMB in the world increases sharply. Theoretically, there is a stress to appreciate the RMB exchange rage. Nevertheless, it is not so easy to do in reality since the government need to consider a lot of influences made by the appreciation of RMB exchange rate.

The appreciation of RMB exchange rate in the short term will bring negative influences to the China's economic growth and employment. It shows concretely as follows:

Firstly, appreciation will restrain exportation growth. After the appreciation of RMB, cost of exportation will increase and international competitiveness of export products will also be weaken, which may go against the continued expansion of exportation. Most of China's export products are with low content of technology. The number of export high-end products with high added value is very small. The appreciation of RMB may increase the nationhood of China, but international competitive advantages of manufacturing industry will lose step by step. Thereby, China's price competitiveness of labour intensive export products will be damaged.

Secondly, appreciation will lead to further expansion of the amount of foreign debt. Since appreciation of RMB attracts large foreign capital to pour into Chinese capital market, the scale of China's foreign debt will enlarge, which will influence the stability of financial market consequentially. In this situation, it is easy to trigger monetary crisis and financial crisis, and easy to cause negative influences for China's economic sustainable development as well.

Thirdly, appreciation will increase employment pressure. Due to the fact that new job opportunities provided by China are mainly from export enterprises and foreign-funded enterprises, the appreciation of RMB will restrain or strike export and finally exacerbate employment situation. Besides, appreciation of RMB will also result in the price growth of non-tradable goods, such as land. Thus, the demand of these non-tradable goods will decrease. And then, there will be further negative influences on the employment market.

Fourthly, appreciation will influence the balance of payments and the stability of financial market. Under the expectation of appreciation of RMB, it will cause cyclic effect which is appreciation-expectation of appreciation-appreciation-expectation of appreciation again. At the present time, there are more and more foreign investments in China. Strong expectation of appreciation of RMB will lead to a quick large inflow of hot money. Once economy fluctuates, this hot money will withdraw quickly, which will in turn result in a huge shock on China's macro economy. Under the global economy, China's economy turmoil will give rise to the unstable factor of the world economy.

Finally, appreciation will strike future economic development badly. Due to the expectation of appreciation of RMB, there will be movements of enclosure in real estate enterprises and a new upsurge of land investments. And then, bubble economy will develop and polarization will be more serious. Similarly, the expectation of appreciation of RMB will attract foreign investors into China's domestic financial market in the short term. However, the negative influences of appreciation of RMB will relieve the attraction of domestic financial market in the future. At last, there will be outflows of foreign funds.

Measurements Taken by Chinese Government

The monetary value of one country is a reflection of its international status on the national economic strength. The change of the monetary value should be matched with its economic strength. Due to the development of Chinese economy, there is a pressure of appreciation on RMB in a medium and long term.

Since the pressure of appreciation is mainly from China's ever-increasing export amount and foreign exchange reserves, the appreciation pressure will be relieved if Chinese government could take the following measurements and make some adjustments.

Firstly, Chinese government may properly lower the tax refund policy for export. The increasing amount of tax refund is equivalent to the depreciation of RMB, and the decreasing amount of tax refund is equivalent to the appreciation of RMB. Taking measurements of adjusting the tax refund policy will counteract the pressure of appreciation on RMB to some extent. At the same time, pressure can also be reduced by exporting more high value-added products.

Secondly, Chinese government may encourage indigenous capitals to invest in foreign countries. In the situation of large foreign capitals pouring into China, Chinese government must expend the outlets of indigenous capitals in order to reduce the pressure of appreciation on RMB. A better choice is to extend capital outflow. Hence, China needs to consider simplifying and extending formalities for examination and approval in order to encourage the strategy of outflow of indigenous capitals.

Thirdly, Chinese government may reform the management of foreign exchange settlement and sale and change the foreign exchange system from compulsive settlement of exchange to aspiration settlement of exchange. This is not only the innovation of foreign exchange settlement system but also the progress of reserve system and inventory policy. Through these measurements, the foreign exchange reserves of corporations and individuals will be increased so that the national currency reserves will be reduced. And then, the pressure of appreciation on RMB will be relieved.

DATA DESCRIPTION AND PRELIMINARY STATISTICS

In this paper, we finally adopt Markov regime switching model with respect to the behaviour of RMB exchange rate middle price against USD from Jan 1994 to Dec 2010.

Before using Markov regime switching model to predict the trends of RMB exchange rate, we need to consider the data that we required in the Markov regime switching model. Since anything cannot be isolated, we consider choosing not only the data of RMB exchange rate itself but also the data of factors which can influence RMB exchange rate. However, the test shows that it is not suitable to use the data of influencing factors. As a consequence, only the data of RMB exchange rate itself is used to predict the trends when we adopt Markov regime switching model.

In this section, there are four parts. The first part is the factor selection, which shows how the influencing factors are selected. The second part is data description. Preliminary statistics is the third part in which different kinds of data are tested to see whether they are appropriate to use or not. In the fourth part, the possible reason why other influencing factors are not suitable for prediction will be described.

Factors Selection

As the aim of this paper is to predict the trends of the exchange rate, the exchange rate is the main research subject. Since USD is the world's principal currency that attracts the most attention, this research adopts the RMB exchange rate middle price against USD.

However, the exchange rate is not an isolated part. It may also be influenced by other factors. As a result, we also need to find out the potential influencing factors and analyse whether these factors influence the exchange rate or not.

Price Level

The change of price level may change the competitive advantages of the import and export products. When other conditions stay constant, the growth of price level may lead to the increase of import and the decrease of export, which will result in the increase of the exchange rate. Conversely, the decrease of the price level may lead to the decrease of import and the increase of export, which will result in the decrease of the exchange rate. Since CPI is used to represent the price level, we choose to use the data of CPI.

Interest Rate

The interest rate may influence the international capital flows. Countries with high interest rate will have capital inflows while countries with low interest rate will have capital outflows. These capital flows will lead to the change of the relationship between demand and supply of foreign exchange market, which thereby influence the exchange rate. In this paper, we adopt deposit rate and lending rate as influencing factors.

Foreign Exchange Reserves

Foreign exchange reserves may also influence RMB exchange rate. If the foreign exchange reserves are ample, RMB exchange rate may increase; if the foreign exchange reserves are lacking, then RMB exchange rate may decrease.

International Gold Price

International gold price may influence RMB exchange rate as well. International gold price represents the world macro-economy to some extent, which may have an influence on RMB exchange rate.

To sum up, we adopt the RMB exchange rate middle price against USD and find out five potential influencing factors, namely CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price.

Data Description

This paper uses one non-linear model which is univariate Markov switching model in terms of the behaviour of RMB exchange rate. From the above part, it can be seen that there are six types of data, which are CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price. According to the data analysis rules, the data we adopt must be in the same period. Consequently, we choose the monthly data in the period from January 1994 to December 2010. Every type of data consists of 204 observations, which are 1224 observations in total. Furthermore, the series are taken in natural logarithms for Markov regime switching model.

Preliminary Statistics

In this part, the relationship between the RMB exchange rate middle price against USD and the influencing factors will be explored by using regression analysis. These influencing factors include CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price. Then, we will pick out the most relevant variables by using correlation analysis. Finally, the regression equation will be test to check its quality. In order to make it convenient, the RMB exchange rate middle price against USD will be abbreviate to the RMB exchange rate in the following content.

The hypothesis and identify correlations between the variables can be tested by using the regression analysis. We will test the following hypothesis and see whether they are true or not:

at least some of the is not equal to 0 (regression insignificant).

In Table 1, the value of R square identifies the portion of the variance in the dependent variable (RMB exchange rate middle price against USD) accounted for by the independent variables. This value reflects that 96.11% of the variance in exchange rate can be predicted from the independent variables which are CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price. It is important to emphasize that R square is only an overall measure of the strength of association and does not indicate the extent to which independent variable is associated with the dependent variable.

Table 1 Model Summary

Model

R Square

Adjusted R Square

R Square Change

Standard Error of the Estimate

1

0.0521

0.0474

55.491

2

0.1429

0.1343

0.0908

52.899

3

0.1433

0.1304

0.0004

53.018

4

0.9583

0.9574

0.815

11.732

5

0.9611

0.9602

0.0028

11.347

In table 2, the p-value is calculated when alpha equals to 0.05. At that level, F-test in ANOVA is significant because the p-value is equal to 0.0000 which is less than 0.001. In this situation, the null hypothesis must be rejected which means that CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price do have effects on the exchange rate.

Table 2 ANOVA

Model

Sum of Squares

Degree of Freedom

Mean Square

F

Significance

1

Regression

34205.0122

1

34205.0122

11.11

0.0010

Residual

622002.545

202

3079.22052

Total

656207.557

203

3232.54954

2

Regression

93757.006

2

46878.503

16.75

0.0000

Residual

562450.551

201

2798.26145

Total

656207.557

203

3232.54954

3

Regression

94031.7751

3

31343.925

11.15

0.0000

Residual

562175.782

200

2810.87891

Total

656207.557

203

3232.54954

4

Regression

628814.933

4

157203.733

1142.04

0.0000

Residual

27392.6248

199

137.651381

Total

656207.557

203

3232.54954

5

Regression

630712.008

5

126142.402

979.63

0.0000

Residual

25495.5495

198

128.765401

Total

656207.557

203

3232.54954

However, knowing the influencing factors is not nearly enough for this research. It is also necessary to understand the significance of each factor. Therefore, we use correlation analysis to explore the significance of each independent variable to the dependent variable.

Table 3 shows the correlation coefficients and these numbers reflect the strength and direction of the linear relationship between the dependent variable and each independent variable. The correlation coefficients of CPI, Deposit Interest Rate and Lending Interest Rate are respectively 0.2283, 0.3592 and 0.3605. These positive numbers indicate that there are positive correlations. In other words, if the value of one independent variable increases, the value of the dependent variable also tends to increase. Obviously, these coefficients are relatively low, which means that these three factors are not very significant for the dependent variable. On the contrary, the correlation coefficients of Foreign Exchange Rate and International Gold Price are respectively -0.9702 and -0.9386, which gives negative correlations. If the value of one independent variable increases, the value of dependent variable has a tendency to decrease. It is easy to see that the absolute values of these two numbers are very close to one, which means that Foreign Exchange Reserves and International Gold Price are very significant to exchange rate.

Table 3 Correlation Coefficients

Exchange rate

CPI

Deposit Interest Rate

Lending Interest Rate

Foreign Exchange Reserves

International Gold Price

Exchange Rate

1.0000

CPI

0.2283

1.0000

Deposit Interest Rate

0.3592

0.8222

1.0000

Lending Interest Rate

003605

0.7557

0.9827

1.0000

Foreign Exchange Reserves

-0.9702

-0.1894

-0.3974

-0.4143

1.0000

International Gold Price

-0.9386

-0.0243

-0.2228

-0.2448

0.9732

1.0000

The results of the correlation analysis above can be validated through scatter plots. Scatter plots show the rough trends that dependent variables vary along with the changes of independent variables. From Figure1 to Figure3, it can be seen that correlation points tend to form along a line going from the lower left to the upper right, which is the same as saying that the correlations are positive. However, the data do not form a line very well, which verify that these three factors are not very significant.

On the contrary, it can be seen from Figure4 and Figure5 that the pattern of dots slopes from upper left to lower right, which is the same as saying that the correlations are negative. Besides, the lines of best fit summarize the trends of these dots quite well, which verify that these factors are very significant.

Figure 1 Correlation between RMB exchange rate and CPI

Figure 2 Correlation between RMB exchange rate and deposit interest rate

Figure 3 Correlation between RMB exchange rate and lending interest rate

Figure 4 Correlation between RMB exchange rate and foreign exchange reserves

Figure 5 Correlation between RMB exchange rate and international gold price

According to the analysis above, foreign exchange reserves and international gold price are the most correlated influencing factors with RMB exchange rate among other factors. Therefore, we adopt these two macroeconomic indicators to study RMB exchange rate. To form a regression equation, we use foreign exchange reserves and international gold price as independent variables and use exchange rate as dependent variable. By using STATA, the P-value of the regression is 0.0000 which is less than0.001. Hence, F-test in ANOVA is significant. Besides, the coefficients of the two independent variables are respectively -0.0077587 and 0.0214754. Consequently, the sample regression equation can be written as follows:

However, before any further study, we must test whether this regression equation is good or not.

Firstly, it is necessary to test whether the residuals () are normally distributed. In this case, we will test the following hypothesis and see whether it is true or not:

: residuals are normally distributed.

: residuals are not normally distributed.

In Table 4, p-value equals 0.3663 which is much more than 0.05. Hence, the null hypothesis is accepted, which means that residuals are normally distributed. From Figure 6, it can be seen directly that the residuals follows a normal distribution.

Table 4 Test of residuals

Variable

Obs

Pr(Skewness)

Pr(Kurtosis)

Adj Chi2(2)

Prob>Chi2

e

204

0.5062

0.2139

2.01

0.3663

Figure 6 Histogram of residuals

Secondly, it is necessary to explore the changes of residuals by plotting the relationship between the residual and the predicted value of y. Because the values of residuals directly reflect the quality of regression equation. As a result, the horizontal axis represents the predicted value of y and the vertical axis represents the residual. It can be seen easily from Figure 7 that the residuals are not distributed evenly around. Moreover, there is obvious divergent trend and outliers. This figure shows that the regression equation we created is not perfect.

Figure 7 Relationship between residuals and predicted values

As a consequence, foreign exchange reserves and international gold price cannot be used to analyse Exchange Rate. In other words, all the five potential influencing factors are not appropriate to use in the Markov regime switching model. As a result, we should only use the data of RMB exchange rate itself to make a prediction of the trends. From Figure 8, the trends of the RMB exchange rate can be seen.

Figure 8 RMB Exchange Rate Middle Price against USD from Jan 1994 to Dec 2010

Possible Reason

The five potential influencing factors which are CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price may have influences on RMB exchange rate in theory. However, the tests we analysed above show that these five influencing factors are not suitable as independent variables. The possible reason of this problem may be the China's exchange rate systems.

There was one reformation of China's exchange rate system on the date of 21 July, 2005. Before this date, the China's exchange rate system is a managed floating exchange rate system which is based on the market supply and demand. After this date, Chinese government became to adopt a managed floating exchange rate system with reference to a basket of currencies which is based on the market supply and demand. Under these two RMB exchange rate systems, the RMB exchange rates are controlled by the Chinese government. This means that the change of RMB exchange rate is made directly by the Chinese government rather than the change of the five potential influencing factors. As a result, the five factors cannot be the independent variables of the RMB exchange rate.

EMPIRICAL MODEL

In this section, we will firstly show the basic knowledge of the univariate Markov regime switching model which is developed by Hamilton (1989). Then, we will use this Markov regime switching model to investigate the regime switching of RMB exchange rate middle price against USD. The prediction of regime switching will be made by Matlab software according to Marcelo Perlin (2011).

Markov Regime Switching Model

A potentially useful method to model nonlinearities in time series is to assume different behaviours (structural breaks) in one subsample (or regime) to another. If the dates of the regime switches are known, modelling can be figured out with dummy variables. For instance, consider the following regression model:

(1)

where, ,

~NID ,

,

,

1 or 2 (Regime 1 or 2).

Normally, it is assumed that the possible difference between the regimes exists because of a mean and volatility shift rather than an autoregressive change. That is:

(2)

where ~NID ,

if is known a priori.

In this case, the problem just becomes a usual dummy variable autoregressive problem.

However, in practice, the prevailing regime is not always directly observable and the transition of regimes is stochastic. Denote as transition probabilities, with. The process in which the future state depends only on the current state rather than the past states is called a Markov process. In other words, in a Markov process the past states are irrelevant for the future state because it doesn't matter how the current state is obtained. The probabilities in a Markov process can be conveniently represented in a matrix form, that is,

,

where means the probability of a switch from state j to state i.

Estimation of the transition probabilities which are represented as is commonly made (numerically) by maximum likelihood as follows. The conditional probability density function for the observations given the state variables St, and the previous observations is:

(3)

where ~ NID .

The chain rule for conditional probabilities is used to yield the following joint probability density function for the variables when past informationis given: . The log-likelihood function to be maximized with respect to the unknown parameters becomes:

(4)

where

and the transition probabilities are:

, and .

Besides, and are called steady state probabilities and given the transition probabilities and, these two probabilities can be obtained as:

, .

RMB Exchange Rate Middle Price against USD Moving Trends Estimation

Consider the following process given by:

(5)

where and follows a normal distribution with zero mean and variance given by . This is the simplest case of a model with a switching dynamic. For the model given in equation (5), the intercept is switching states given an indicator variable. In other words, if there are states, there will be values for and.

In this research, we assume that there are two states in the exchange rate. As a result, an alternative representation is:

for state 1 (6)

for state 2 (7)

where ~ for state 1 (8)

~ for state 2 (9)

This representation clearly shows two different processes for the dependent variable which is the exchange rate. When the state of the exchange rate for time is 1 (or 2), then the expectation of the exchange rate is (or) and the volatility of the exchange rate is (or). These two volatilities ( and) indicate the higher uncertainty regarding the predictive power of the model in each states of the exchange rate.

EMPIRICAL RESULTS AND ANALYSIS

In this section, there are two parts. The first part will present the results of the prediction by using the Matlab software. Then, the second part will make an analysis of the empirical results and see whether Markov regime switching model can forecast the moving trends of RMB exchange rate well or not.

Empirical Results

By using the Matlab software, we get the empirical results of the moving trends estimation of exchange rate middle price against USD, which are shown in Table 5. The results indicates that the likelihood ratio test for the null hypothesis of linearity is statistically significant, which means that linearity is strongly rejected.

Table 5 RMB Exchange Rate Moving Trends Estimation by MRS

State

Parameter

Estimate

Standard Error

Expected Duration

State 1

-0.0001

0.0001

56.05

0.0010

0.0001

time periods

State 2

-0.0126

0.0064

23.99

0.0544

0.0040

Time periods

Transition Probabilities

P (regime 1)

0.98

q (regime 0)

0.96

Final log Likelihood

689.0672

Next, we will pay close attention to some extremely important points displayed in the results. Firstly, the expected value of the switching variable at state 1 which is represented as is -0.0001 while the expected value of the switching variable at state 2 which is represented as is -0.0126. These two outcomes indicate the expected values of the mean growth rate. Since these two numbers ate both negative, this implies that regime 1 and regime 2 are both downward regime. In other words, RMB exchange rate middle price against USD from Jan 1994 to Dec 2010 is appreciated in not only regime 1 but also regime 2. Moreover, the absolute value of u1 is much larger than that of u2. This represents that even though decreases of RMB exchange rate happen in both regimes, the RMB exchange rate against USD in regime 2 goes down much faster than the RMB exchange rate against USD in regime 1. Secondly, the standard deviation of this model which is represented as is 0.0010 while the standard deviation of this model which is represented as is 0.0544. Compared to the first standard deviation, the second standard deviation is much larger. This comparison of the standard deviation means that the RMB exchange rate in regime 1 is more volatile than that in regime 2. Thirdly, the expected durations of these two regimes are 56.05 months and 23.99 months respectively. It can be easily seen that regime 1 lasts longer than regime 2. Finally, transition probabilities which are represented as are presented through the results. This matrix suggests that the probability of staying in regime 1 is 0.98 while the probability of staying in regime 2 is 0.96. In addition, this matrix also implies that the probability of a switch from regime 2 to regime 1 is 0.04 while the probability of a switch from regime 1 to regime 2 is 0.02. In summary, the RMB exchange rate against USD in regime 1 is relatively stable while the RMB exchange rate against USD in regime 2 with a higher volatility decreases faster.

Figure 8 and Figure 9 show the smoothed states probabilities lines and filtered states probabilities lines of being in regime 1 and regime 2 respectively along the RMB exchange rate middle price against USD. It can be seen that the sum of the states probabilities of state1 and state 2 in both figures is equal to one, which is in keeping with the fact in reality that the trends of RMB exchange rate follow either regime 1 or regime 2. Another similarity of these two figures is that RMB exchange rate middle price against USD decreases frequently during the period from the year 1994 to the year 1997 and the period from the year 2005 to the year 2010. 两幅图的区别.

Figure 9 Smoothed States Probabilities (Moving Trends)

Figure 10 Filtered States Probabilities (Moving Trends)

Results Analysis

The outcomes of the moving trends estimation by Markov regime switching are in good agreement with the reality.

From Jan 1st of the year 1994, China began to carry out a merging of RMB exchange rates, which means that the official RMB exchange rate and the regulatory market RMB exchange rate were unified. From then on, China have adopted a single, managed floating exchange rate system which is based on the market supply and demand. During the period from the year 1994 to the year 1997, the RMB exchange rate was declining gradually. The RMB exchange rate middle price against USD went down from the value of 870 to the approximate value of 829.

In July of the year 1997, the Asian financial crisis broke out. Firstly in Thailand, it was forced to float the Thai Baht, which promptly plunged. Soon after, the Asian financial storm swept through Malaysia, Singapore, Japan, Korea and so on. This financial crisis broke the scenario of the Asian's rapid economic development and depressed the economy. Even though the currencies of many countries in Asian depreciated substantially, Chinese government seriously took its responsibility to the national and international economy to make a decision that RMB would not be depreciated. This behaviour played a very important role in the economic stability and development of Asian or even the world. As a result, the RMB exchange rate remained steady. The value of the RMB exchange rate middle price against USD is about 827 till the year of 2005.

On July 21 of the year 2005, China adopted a new RMB exchange rate system. This system is a managed floating exchange rate system with reference to a basket of currencies which is based on the market supply and demand. After this change of the RMB exchange rate system, RMB appreciated step by step. At the end of 2010, the RMB exchange rate middle price against USD is about 665.

CONCLUSION AND DISCUSSION

In this paper, we have applied the Markov regime switching model to the RMB exchange rate. Firstly, we try to use five influencing factors which are CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price to study the trends of RMB exchange rate. We choose the monthly data of the RMB exchange rate middle price against USD, CPI, deposit interest rate, lending interest rate, foreign exchange reserves and international gold price from Jan 1994 to Dec 2010. However, after using regression analysis, correlation analysis and some other tests by stata, it can be found that using these influencing factors to study the trends of RMB exchange rate is not appropriate. This result may be due to the macroeconomic coercive control of Chinese government. Afterwards, we use the monthly data of RMB exchange rate itself to predict the trends by using matlab successfully. The outcomes show that there were two regimes during the period from Jan 1994 to Dec 2010. Even though these two regimes are both downward, the first regime is relatively stable with duration of 56.05 months while the second regime has a higher negative growth rate and a higher volatility with duration of 23.99 months. The results indicate that the moving trends forecasted by Markov regime switching model are quite matched with the reality.

According to the trends prediction we have made, the RMB exchange rate will continue to decrease in 2011. This forecast is based on the following analysis. Firstly, the year 2011 is the beginning year of China's 12th Five-Year Plan. As China's monetary policy carries out changes from accommodative monetary policy to steady monetary policy, the probability of increasing the interest rate will be raised. The increase of the interest rate difference between China and other countries will aggravate the inflows of international capital, which will lead to the pressure increase of RMB appreciation. Secondly, as the inflationary pressure grows, appreciation of RMB will offset the imported inflationary pressure which is resulted from the price increase of the global food, international energy and raw materials. Thirdly, some countries, such as America, have not been recovered from the financial crisis, which will also lead to the decrease of RMB exchange rate against USD.

This paper has already studied the moving trends of RMB exchange rate by using Markov regime switching model. However, Markov regime switching model is not the only method of studying the RMB exchange rate. In the future, we may choose another method, such as artificial neural network, to study the RMB exchange rate. Besides, we can also compare the results from these two different methods to see which research method is better.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.