## Test For Stationarity Unit Root Test Finance Essay

The unit root trial is conducted before an analysis of cointegration. Unit root trial is based on proving the void hypothesis against the alternate hypothesis of stationarity. Several surveies have demonstrated that macroeconomic clip series are non-stationary. In line with the conditions of non-stationarity informations, the normal belongingss of Durbin-Watson ( DW ) , t statistics and the step of R2 will no longer keep and this will take to specious consequences in instance we perform the arrested development. In order to prove for stationarity, the Augmented Dicker Fuller ( ADF ) trial will be employed.

## Augmented Dicker Fuller Test

The ADF trial is used to look into the relationship between dependant and independent variables in the long tally. An ADF theoretical account can be shown as follows:

Where ?t is a white noise mistake term, Yt = { LSEDX, LM2, LEXC, LYIELD, LOIL, LCPI } which is the series being tested, a?† is the difference operator and T is the clip tendency.

The hypothesis under the ADF trial:

H0: ? = 0

H1: ? & A ; lt ; 0

The void hypothesis is that ? = 0. The clip series is stationary when the void hypothesis is rejected. The void hypothesis of ? = 0 will be tested against the alternate hypothesis ? & A ; lt ; 0. If the series is non stationary at flat signifier, it will be differenced 500 times to be stationary to find the order of integrating.

## Cointegration Trial

Cointegration is employed to cognize whether the theoretical account has cointegrating vectors or non. In other words cointegration examines the long-term relationship between macroeconomic variables and stock returns. In most instances if two variables are integrated of order one, that is I ( 1 ) , their additive combination will besides be stationary. Granger ( 1986 ) stated that a cointegration trial acts as a pre-test to avoid specious arrested development fortunes. For the cointegration analysis, the Maximum Likelihood appraisal method of the Johansen and Juselius ( 1990 ) multivariate attack will be used. This trial stipulates that the rank of the matrix is equal to the figure of cointegrating vectors and it besides determine the long tally relationship between the explained and explanatory variables.

## Johansen and Juselius Procedure

The Johansen method is a process for proving cointegration of several variables which are integrated of order one. This trial allows more than one cointegrating relationship.

Assume that a set of g variables are I ( 1 ) and are said to be cointegrated. It is possible to build a VAR including K slowdowns holding these variables.

Yt = ?1 yt-1 + ?2 yt-2 + … + ?k yt-k + Greenwich Mean Time

Where,

Yt is g x 1, matrix of non-stationarity.

?i is g x g matrix of parametric quantities.

In order to use the Johansen trial, the VAR should be reformulated into a Vector Error Correction Model ( VECM ) as shown below:

a?†yt = ?“1a?†yt-1 + ?“2a?†yt-2 +…+?“k-1a?†yt-k+1 + ? yt-k + Greenwich Mean Time

Where,

?“i = – ( I – ?1-…-?i ) , I = 1, … , k-1 and ? = – ( I – ?1 -…-?k ) .

Information about the short and long-term accommodation to alterations in yt via estimations of ?“ and ? severally is specified in the system. This VAR contains g variables in the first differenced signifier on the left manus side and k-1 slowdowns of the dependent variables ( differences ) on the RHS, each with a ?“ connected to it. The Johansen trial is really sensitive to the chosen length slowdown ; therefore it is of import first to find the appropriate slowdown length before utilizing proving for cointegration.

The Johansen trial focuses on the scrutiny of the ? matrix. It can be defined as the long tally matrix because in equilibrium, all a?†yt-i will be zero and by doing the mistake footings ut, equal to their respective expected value of nothing, we will acquire ? yt-k = 0.

? = ?? ‘ , where ? denotes the velocity of accommodation to disequilibrium and ? is a matrix of long tally coefficients.

Since a?†yt…a?†yt-k+1 are all I ( 0 ) but yt is I ( 1 ) while ? yt-k must be stationary for ut to be I ( 0 ) , there are three possible instances where the status ? yt-k ~ I ( 0 ) can be satisfied:

When are variables in the system are I ( 0 ) which means there is full cointegration, that is Rank ( ? ) = p. A simple VAR is the suited theoretical account in this instance.

When Rank ( ? ) = 0 connoting no cointegration among the variables in the long tally. The appropriate theoretical account will be a first difference VAR.

The last manner for the status to be satisfied is when ? has a reduced rank of R ? ( n-1 ) . In other words, there exists up to ( n-1 ) cointegrating relationships: ? ‘ yt-k ~ I ( 0 ) . In this instance the being of R ? ( n-1 ) cointegrating vectors can be noted in ? with ( n-r ) non-stationary vectors. Merely the cointegrating vector in ? enter the VECM implying that the last ( n-r ) are efficaciously zero

Johansen and Juselius ( 1990 ) stipulate two trial statistics for cointegration which can be shown in equation ( I ) and ( two ) :

Trace Trial: Ttrace =

Where, T is the entire figure of observation, N is the figure of variables and Rhode Island is the i-th brace of variables. T hint has a chi-square distribution with N-r grades of freedom. Large value of T hint gives grounds against the hypothesis of R or fewer cointegration vectors.

Maximum Eigenvalue Test: Tmax = -T ln ( 1 – ?r+1 )

The maximal eigenvalue trial evaluates the void hypothesis of H0: R = r0 against H1: R = r0+1. The void hypothesis of R cointegration vectors id tested against the option of R + 1 cointegrating vectors.

## Choice of the slowdown length

Before continuing to the Johansen process, the finding of the slowdown length of the VECM is of import. The slowdown length should be little plenty in order to let appraisal and high plenty to vouch that the mistakes are about white noise. An deficient slowdown length can ensue in the rejection of the void hypothesis of no cointegration and on the other manus over-parameterization of the slowdown length may take to loss of grades of freedom. The Johansen trial employs the information multivariate standards viz. the Akaike information standard ( AIC ) and Schwarz Bayesian information standard ( SBC ) .

The Akaike information standard is a gage of the goodness of tantrum of a theoretical account. It emphasizes the trade-off between prejudice and discrepancy in the theoretical account building, or that of truth and complexness of the theoretical account. It besides allows for comparing in the theoretical account.

The AIC is defined as

AIC= 2K-2ln ( L )

Where K is the figure of parametric quantities in the statistical theoretical account and L is the maximized value of the likeliness map for the estimated theoretical account.

The Schwarz Bayesian standard is a standard for the theoretical account choice among a set of parametric theoretical accounts with different Numberss of parametric quantities. The job of over fitting arises while gauging ; hence SBC solves this restraint through a punishment ( larger than the AIC ) term for the figure of parametric quantities in the theoretical account.

The expression for SBC is as follows:

SBIC= -2Ln L+ KLn ( N )

Where K = no of parametric quantities, n is the figure of informations points in the ascertained informations and L is the maximized value of the likeliness map for the theoretical account under appraisal.

## Vector Error Correction Model

An mistake rectification theoretical account is a dynamic system with the features that the divergence of the current province from its long-term relationship will be fed into its short-term kineticss. The VECM is a full information upper limit likeliness appraisal theoretical account and it yields more efficient calculators of cointegrating vectors. Without necessitating a particular variable to be normalized, the VECM enables the proving for cointegration in a whole system in one measure.

The general signifier of a VECM is shown below:

a?†Yt =

Where, and are the constituents of the vector autoregression in first differences and mistake rectification constituents. Yt represents a P ten 1 vector of variables and is integrated of order one. ? is a p ten 1 vector of invariables, K is the slowdown construction and Greenwich Mean Time is a p ten 1 vector of white noise mistake footings. Short tally parametric quantities are represented by ?“j which is a p ten P matrix of these coefficients across p equations at the jth slowdown. ? is a P x R matrix of cointegrating vectors and ? represents the accommodation parametric quantities which is the velocity of mistake rectification mechanism. a?† is the difference operator.

## Granger Causality Test

Granger causality trial is employed to analyse the causality way between stock market index and its selected determiners. The causality trial requires that all informations series into observation should be stationary. This trial is carried out to see the short-term causality running from independent variables to the dependant variable.

The Chi Wald trial is used to analyze the causality between the dependant variable and the explanatory determiners. The hypothesis is shown below:

H0: One variable farmer does non granger do the other

H1: One variable farmer cause the other

The determination regulation is that if the qi statistic is less than the chi critical value, we do non reject H0 and if the qi statistic is greater than the chi critical value we reject H0.

## Chapter 5: Analysis AND FINDINGS

Table 1: Descriptive Statisticss: June 1998 to June 2010

## Variables

## Observations

## Mean

## Std divergence

## Minimum

## Maximum

LSEDX

145

6.558

0.572

5.831

7.603

LM2

145

25.719

0.586

24.972

30.425

LEXC

145

3.373

0.0881

3.195

3.544

LYIELD

145

2.135

0.335

1.247

2.575

LOIL

145

3.659

0.596

2.343

4.887

LCPI

145

4.769

0.0748

4.636

4.933

a?† LSEDX

145

0.00877

0.053

-0.195

0.226

a?† LM2

145

0.01006

0.544

-4.587

4.609

a?† LEXC

145

0.00208

0.022

-0.081

0.117

a?† LYIELD

145

-0.00666

0.065

-0.197

0.232

a?† LOIL

145

0.01242

0.091

-0.311

0.201

a?† CPI

145

0.00062

0.034

-0.291

0.029

( Beginning: Computed )

The above tabular array depicts the descriptive statistics of the variables. As can be seen the standard divergence of Semdex is 0.572 which indicates that the Stock Exchange of Mauritius is rather a volatile market. Furthermore the mean monthly return of semdex shows a just norm of 0.87 % which is equal to an annualized return of 10.44 % per twelvemonth with a standard divergence of 5.3 % . The Semdex earns maximal return of 22.6 % in one month and maximal loss of 19.5 % .

## The Unit Root Test

The first measure of the analysis is to look into the presence or absence of stationarity in the time-series informations by utilizing the Augmented Dicker Fuller ( ADF ) trial. The appropriate slowdown length for each variable is determined prior to executing the trial. As a affair of fact, for cointegration analysis to be valid, the clip series informations should be integrated of the same order. In most instances the series must be integrated of order one. The tabular arraies below summarize the consequences of the ADF trial.

Table 2: Unit of measurement root trial consequences at flat signifier

## Seriess

## ADF Test Statistic

## Lag-Length

## p-value

## Decision

LSEDX

-0.648

4

0.8547

Non-Stationary

LM2

-1.839

4

0.3615

Non-Stationary

LEXC

-1.958

1

0.3055

Non-Stationary

LYIELD

-0.772

2

0.8273

Non-Stationary

LOIL

-1.851

2

0.3557

Non-Stationary

LCPI

-2.950

1

0.0398

Non-Stationary

( Beginning: computed )

Table 3: Unit of measurement root trial consequences at first difference

## Seriess

## ADF Test Statistic

## Lag-Length

## p-value

## Decision

LSEDX

-4.118

3

0.0009

I ( 1 )

LM2

-8.848

4

0.0000

I ( 1 )

LEXC

-7.760

1

0.0000

I ( 1 )

LYIELD

-5.663

2

0.0000

I ( 1 )

LOIL

-5.713

2

0.0000

I ( 1 )

LCPI

-8.500

1

0.0000

I ( 1 )

( Beginning: computed )

The consequences show that at flat signifier all variables are non-stationary since their several test-statistic does non transcend its critical value. Therefore the void hypothesis of a unit root can non be rejected and any standard arrested development will bring forth specious consequences. However, when the first difference was done, all variables were found to be stationary. Stock market index, money supply, exchange rate, overall leaden output on exchequer measures, oil monetary values are found to be stationary at 5 % important degree and consumer monetary value index at 1 % important degree. This indicates that all variables are integrated of order one. We can state that there is a possibility of a long tally relationship among the variables. In order to happen such a relationship we proceed to the Johansen Multivariate trial.

## Cointegration Test Results

Before continuing to the cointegration trial, it is of import to find the optimum slowdown length. The theoretical account slowdown choice was determined by the Schwarz Information Criterion ( SIC ) and the Akaike ( AIC ) Information Criterion. An optimum slowdown length is critical because it avoids the job of consecutive correlativity. The aim will be to choose the figure of parametric quantities which minimizes the information standard. The drawback of SIC is that it tends to undervalue the slowdown order, while adding more slowdowns will increase the loss of grade of freedom. The AIC is chosen as a taking index in order to do certain that there is no staying autocorrelation in the VAR theoretical account. Since monthly frequences are used in the survey, Ivanov and Kilian ( 2001 ) mentioned that AIC tends to be more accurate with monthly informations. Therefore the optimum slowdown length selected is 2. The tabular array below shows the consequences for the choice of the appropriate slowdown length

Table 4: VAR slowdown order choice by information standards

## Slowdown

## AIC

## SBIC

0

-2.04432

-1.91884

1

-15.3857

## -14.5073*

2

## -15.4818*

-13.8506

3

-15.37

-12.9859

4

-15.041

-11.904

( Beginning: computed )

The Johansen and Juselius ( 1990 ) process is utile in finding whether there exists a long-term relationship among the variables. Table 5 estimates the figure of cointegrating vectors which exists between stock market index and the macroeconomic variables.

Table 5: Consequences from Johansen ‘s Cointegration Test ( Trace and Maximum Eigenvalue )

## Trace Test

Null hypothesis

Alternate Hypothesis

Eigenvalue

Trace Statistic

5 % critical value

R = 0

R = 1

126.7153

94.15

R ? 1

R = 2

0.38030

58.2865*

68.52

R ? 2

R = 3

0.17259

31.1943

47.21

R ? 3

R = 4

0.10633

15.1178

29.68

R ? 4

R = 5

0.06635

5.3007

15.41

## Max-Eigen Trial

Null Hypothesis

Alternate hypothesis

Eigenvalue

Max-Eigen Statistic

5 % critical value

R = 0

R = 1

68.4287

39.37

R ? 1

R = 2

0.38030

27.0923

33.46

R ? 2

R = 3

0.17259

16.0765

27.07

R ? 3

R = 4

0.10633

9.8170

20.97

R ? 4

R = 5

0.06635

5.2052

14.07

( Beginning: computed )

As can be observed in the tabular array above, both the Trace statistic and Max-Eigen statistic indicate the presence of one cointegrating vector. In other words the series has a cointegrating rank of one. We reject the void hypothesis of no cointegrating equilibrium at the 5 % significance degree and conclude that there is a long tally relationship between stock market returns and macroeconomic basicss. The Johansen Test suggests two chief averments. First, in the long tally the variables move together and short term divergences will be corrected towards equilibrium. Second, when there is cointegration this indicates causality in at least one way

Since one cointegrating vector has been found, the relationship between stock market index and the macroeconomic variables can non be modeled in a VAR. In this instance, a VECM is the most appropriate theoretical account. Thus the cointegrating relationship is tested through a VECM to analyze the long tally relationship.

The Johansen trial showed that there exists a long tally relationship between the variables. The cointegrating vector ( normalized on stock market index ) represents the long tally relationship between stock market and the factors, viz. money supply, exchange rate, overall leaden output on exchequer measures, oil monetary values and consumer monetary value index is given by:

Crosstalk = ( LSEDX LM2 LEXC LYIELD LOIL LCPI C ) .

? = 1.000 -4.411 10.21 -1.068 1.024 -3.648 88.40

The coefficients in the equation are long-run snap steps because the variables have been transformed into logarithmic signifier. Thus the cointegration relationship can be re-expressed in table 6 below.

Table 6: Long tally cointegrating relationship between the variables

Dependent variable ( LSEDX )

Independent Variables

LM2

LEXC

LYIELD

LOIL

LCPI

Changeless

coefficient

4.411**

-10.21**

1.068**

-1.024**

3.64

88.40

p-value

0.000

0.000

0.050

0.006

0.089

Note: ** denotes significance at 5 % degree severally.

## LSEDX = 4.411LM2 – 10.21LEXC + 1.068LYIELD – 1.024LOIL + 3.648LCPI + 88.40

From the above equation, it can be observed that money supply ( M2 ) has a positive relationship with the stock market index. A 1 % rise in money supply triggers a 4.41 % addition in stock returns. The consequence supports the findings of Mukherjee and Naka ( 1995 ) for Japan and Ratanapakorn and Sharma ( 2007 ) for the United States. This positive relationship can be explained by the portfolio permutation theoretical account. An addition in money supply in the fiscal system will convey about portfolio rebalancing with other assets including securities. Investors will switch their financess from non-interest bearing pecuniary assets to equities. Another intuitive account is that pecuniary growing creates liquidness and as a consequence involvement rates will fall due to extra liquidness. Cost of borrowing diminutions, investing and aggregative demand addition. As a effect hereafter hard currency flows of corporates rise. In line with the dividend price reduction theoretical account, portion monetary values and finally returns will travel up. The determination indicates that pecuniary policies influence mostly the stock market in Mauritius. Money supply is a important factor in the long tally and therefore it is the chief drive force of the stock market in Mauritius.

The long tally association between exchange rate and stock market index shows a negative value. If the rupee depreciates by 100 footing point, stock returns will fall by 10.21 % . This negative relationship can be explained by the position of Mauritius in international trade. Since Mauritius is a net importer, a depreciation of the rupee vis- & A ; agrave ; -vis the dollar will increase the production costs of houses which import natural stuffs and goods from abroad. Profit border of those companies will fall and therefore their portion monetary values listed in the stock market will diminish. The determination is in line with Maysami and Koh ( 2000 ) for Singapore. The portfolio balance theoretical account is another manner of explicating this opposite relationship. Harmonizing to Ibrahim and Aziz ( 2003 ) , we should see as an investor ‘s point of position. A depreciation will drive portfolio investing out of the state. Most investors will retreat weak currencies from their portfolio. There will be capital flight and this will impact adversely the net incomes of companies due to loss of capital from foreign investors. As a consequence portion monetary values and returns of those companies will fall. Since exchange rate is a important factor in the long tally, any alteration in the dollar/rupee relationship will do the stock market injury. It can be concluded that exchange rate is the most terrible hindrance for the Mauritanian stock market.

Treasury measures seem to be an of import pricing factor in the long tally. In line with Ratanapakorn and Sharma ( 2007 ) , there is a positive relationship between stock market index and exchequer measures. A 1 % rise in the exchequer measures ‘ rates will bring on a 0.87 % rise in stock returns. This suggests that investors do non see exchequer measures as an alternate investing chance. They like to unite both equities and riskless assets like exchequer measures in their portfolios. Thus additions in exchequer measures ‘ rates lead to a rise in investing in stocks doing stock returns to increase.

There is a negative relationship between petroleum oil monetary values and stock market index. This is non surprising since Mauritius is a net oil importer. A 1 % rise in oil monetary values will do stock returns to fall by 1.024 % . From the consequences it can be seen that oil monetary value is a important factor in finding stock returns. An addition in the monetary value of oil is a major drawback for houses which use oil as input. Those houses will happen it hard to absorb the extra addition in the cost of production. This will take to a autumn in the general degree of economic activity and therefore net incomes will diminish and future hard currency flows will be lower. Harmonizing to the dividend price reduction theoretical account a autumn in hard currency flows will do a diminution in stock monetary values.

In line with the findings of Nasseh and Strauss ( 2000 ) for the six European states, we have found a positive relationship between consumer monetary value index ( CPI ) and stock market returns. Stock returns rises by about 3.64 % if CPI additions by 1 % . Although this consequence contradicts with others which have found a negative relationship, another statement can be given for this positive relationship. A rise in the monetary values of concluding goods at a rate higher than input monetary values will augment the overall hard currency flows of houses. This will increase the portion monetary values of those houses. Furthermore portion monetary values should associate positively with rising prices via the hedge operation. Equity serves as a protection against the menace of rising prices. The higher the rising prices rate, the higher the demand of a peculiar portion. Therefore we conclude that stocks in the SEM can be used partially or to the full as a hedge against rising prices.

Table 7: Coefficient of Error Correction Footings

## Variables

## a?†LSEDX

## a?†LM2

## a?†LEXC

## a?†LYIELD

## a?†LOIL

## a?†LCPI

ECT t-1

## 0.006181

0.0229

-0.00219

-0.00373

0.000158

0.000153

p-value

0.085**

0.000**

0.173

0.359

0.980

0.951

Note: ** denote significance at 5 % degree

The positive mistake rectification term is important at 5 % degree. The mistake rectification term induces a positive alteration in stock returns back towards equilibrium. Stock market returns adjust in order to extinguish about 0.06 % of a unit alteration in the divergence from the equilibrium relationships created by alterations in the macroeconomic variables. The mistake rectification coefficients of all other variables are little which mean that there is a deficiency of rapid constitutional accommodation to the long-term equilibrium.

Next we examine the specification of the theoretical account. First and foremost we investigate the stableness of the VECM. Figure 1-I in appendix shows the characteristic root of a square matrixs of the comrade matrix of VECM. As can be seen about all the roots have modulus less than one and lie inside the unit circle, we conclude that the estimated VECM is stable. Second we check for consecutive correlativity in the remainders. We employ the Lagrange-Multiplier trial and the consequences suggest no job of consecutive correlativity.

Table 8: Lagrange-Multiplier Trial

## Slowdown

## Chi2

## df

## Prob & A ; gt ; Chi2

1

41.5463

36

0.24181

2

33.6253

36

0.58206

H0: No autocorrelation at slowdown order

## Granger Causality Test

Since there is a long tally relationship between stock market index and macroeconomic variables, we can analyze the short tally kineticss utilizing the Granger-Causality trial. Harmonizing to Granger ( 1986 ) , if a brace of I ( 1 ) series are cointegrated, there exists a unidirectional causal relationship in either manner. The Table below shows the consequences of the Granger-Causality trial.

Table 9: Granger Causality Chi-Wald Trials

## Null Hypothesis

## Chi-Square Statistic

## Prob & A ; gt ; Chi Square

## Inference

LSEDX does non granger cause LM2

6.2243

0.045

Causality

LM2 does non granger cause LSEDX

3.0344

0.219

NO CAUSALITY

LSEDX does non granger cause LEXC

3.6567

0.161

NO CAUSALITY

LEXC does non granger cause LSEDX

6.5176

0.038

Causality

LSEDX does non granger cause LYIELD

3.0291

0.220

NO CAUSALITY

LYIELD does non granger cause LSEDX

2.5822

0.275

NO CAUSALITY

LSEDX does non granger cause LOIL

2.2127

0.331

NO CAUSALITY

LOIL does non granger cause LSEDX

5.8686

0.053

Causality

LSEDX does non granger cause LCPI

0.04737

0.997

NO CAUSALITY

LCPI does non granger cause LSEDX

0.03571

0.728

NO CAUSALITY

( Beginning: computed )

Figure 1: Causality Relationship Direction

LSEDX Money Supply

Exchange Rate LSEDX

Oil Price LSEDX

Note: denotes one manner causality

As can be seen in table 9, there exist no farmer causality between exchequer measures, consumer monetary value index and stock market returns as it is undistinguished at 5 % degree. Therefore the void hypothesis of causality is rejected. On the other manus one-way causality can be depicted between money supply, exchange rate, oil monetary values and stock returns as the chi-square statistics are important at 5 % degree.

Stock market will causal money supply. This is in line with Chakravaty ( 2007 ) who found a unidirectional relationship between stock market and money supply. Money nowadays in the economic system is determined to a great extent by the stock market. A roar in the stock market shows that the economic system is making good and will be comfortable in the hereafter. Household wealth will increase and people will prefer to pass more. They will put their money in more liquid assets since the stock market in making good. At the same clip houses will set about capital undertakings since their assurance on the stock market has risen. This will bring on a positive demand daze to the economic system and hence aggregative money supply will increase. Semdex can be considered a prima index of money supply in Mauritius.

The unidirectional causality between stock market and exchange rate can be attributed to the openness of Mauritius to international trade. The state is more engaged in the exportation and importing of goods. The assets and liabilities of companies involve in international trade are more open to the volatility of exchange rate in the foreign exchange market. Any alteration in the par value of the exchange rate will impact on the profitableness of the companies. For case an grasp of the rupee will do exports more expensive and therefore this will impact the hard currency flows of the exporting houses. Consequently their portion monetary values will fall in the stock market. In Mauritius companies in EPZ sector is more open to volatility in the exchange rate.

Oil monetary values farmer cause stock market returns. This consequence is consistent with Heista ( 2011 ) . The latter found the same consequence for Mauritius. It is by and large believed that there is a negative correlativity between oil monetary values and stock returns. Since Mauritius is a major importer, many companies prefer to freight their goods which include seafood, fabrics, dairy merchandises, other indispensable trade goods and natural stuffs. Oil is the premier factor in the cost of transporting. Rising oil monetary values means higher transit and production costs for those companies. First, corporate net incomes will fall due to the addition in the cost of production and this will impact of the portion monetary values. Second, the addition in cost of production is translated in higher monetary values for consumers. Peoples will pass sagely and therefore consumer disbursement will fall. They will non set about investing activities since their buying power has fallen. As a consequence stock market will non make good and returns will diminish.

## Chapter Six: CONCLUSION AND POLICY RECOMMENDATIONS

## Decision

Using the Johansen multivariate cointegration analysis on a monthly clip series for the period 1998 to 2010, the survey investigates whether macroeconomics variables affects stock markets returns in Mauritius. Money supply, exchange rate, oil monetary values, exchequer measures, consumer monetary value index were the economic forces while stock market index was used as a placeholder for stock returns in the analysis.

The empirical consequences showed the being of a long relationship between those variables and stock returns. This demonstrates the presence of macro-information in the Stock Exchange of Mauritius. Harmonizing to the VECM, there exists a positive relationship between money supply, exchequer measures, consumer monetary value index and stock returns. On the other manus exchange rate and oil monetary values affect negatively the returns of semdex. Furthermore money supply is found to be the chief drive force of the SEM while exchange rate is its most terrible hindrance. This survey has serious deductions for fiscal analysts, portfolio directors and policymakers.

This thesis farther analyzes the short term causal relationships that exist among the variables by using the Granger Causality trial. Consequences showed three unidirectional causal relationships. Stock returns farmer causes money supply while exchange rate and oil monetary values farmer causes stock returns. We therefore demonstrate that by utilizing macroeconomic variables, the behavior of stock monetary values can be predicted in Mauritius. The presence of cointegration illustrates that the SEM does non look to be efficient because macroeconomic variables can function as a tool to calculate stock returns.

## Policy Recommendations

Since there is a positive relationship between rising prices and stock returns in the long tally, the listed companies should take enterprises to do their stocks attractive to investors because equities seem to be a good hedge against rising prices. Sustainable and profitable undertakings should be undertaken by the listed houses in order to hike their fiscal public presentation over the long term. Consequently investors will set their money on stocks which have the possible to appreciate coupled with the fact that portion returns rises as rising prices goes up, the portions can be the appropriate assets for investors to fudge against rising prices.

The Central Bank should closely supervise the development of the Rs/US $ exchange rate on a regular footing. Any deformation in the rate will do injury to the stock market since exchange rate affects negatively stock returns. Export policies should be implemented so as to hike the volume of exports in order to better the public presentation of the balance of trade. A significant balance of payment shortage will impact on the exchange rate and cause injury to the stock market. The pecuniary authorization should be cautious in its aim to keep monetary value stableness. Any alterations in the money supply will impact on the stock market and capital formation ensuing in a diminution in the economic system.

## Restrictions of the Study

This survey has employed merely five macroeconomic variables to measure its influence on the stock market. Variables such as GDP, foreign direct investing should be employed to lend farther to the bing literature. Structural interruptions during period of crisis can be included to analyze its deductions on stock market public presentation.

## Scope for farther research

Using the same process, we can compare SEM with emerging markets in Sub Saharan Africa or with states in East Asia. Furthermore other surveies can look into on the influence of long established stock markets on the SEM in order to cognize whether SEM is cointegrated with those markets.