Econometrics
Content:
- Describing of economic system. Main Problems. Review of modern scientific articles.
- Collection of statistical data
2.1. Determination of the endogenous and exogenous variables.
2.2. Describing of statistical data related to the model.
- 3. Econometric model
3.1. Model specification
3.2. Steps of estimating the coefficients of the model
3.3. Specification of estimating model
3.4. Interpretation of the model
3.5. Tests: R^2, F-test, T-test, GQ-test, DW-test
3.6. Construction of the confidence interval
3.7. Checking model adequacy
- Economic analysis of model results
- Model forecasting
- Conclusions
- Recommendations
- References
- Appendix
- Theoretical base of the work
Malaysia is a highly open, upper-middle income economic system. Malaysia was one of 13 countries identified by the Commission on Growth and Development in its 2008 Growth Report to have recorded average growth of more than 7 percent per year for 25 years or more. Economic growth was inclusive, as Malaysia also succeeded in nearly eradicating poverty.
From an economy dominated by the production of raw natural resource materials, such as tin and rubber, even as recently as the 1970s, Malaysia today has a diversified economy and has become a leading exporter of electrical appliances, electronic parts and components, palm oil, and natural gas. After the Asian financial crisis of 1997-1998, Malaysia continued to post solid growth rates, averaging 5.5 percent per year from 2000-2008. Malaysia was hit by the Global Financial Crisis in 2009 but recovered rapidly, posting growth rates averaging 5.7 percent since 2010.
Though extreme poverty is less than 1 percent, pockets of poverty remain and income inequality remains high relative to other developed countries: Malaysia’s gini coefficient of income inequality stood at 0.41 in 2014, compared with 0.31 and 0.33 in the Republic of Korea and Japan (both as of 2010), for example. Real income of the bottom 40 percent of households increased by an average 6.3 percent per year between 2009 and 2012, compared to 5.2 percent for the average household, suggesting the benefits from growth were being shared.
Malaysia’s near-term economic outlook remains broadly favorable, reflecting a well-diversified economy, despite some risks. The Government has taken steps to broaden the revenue base, in particular by introducing a Goods and Services Tax in 2015 and by removing fuel subsidies in 2014. Recent increases in the minimum wage and public sector salaries to support households’ income may prove challenging to sustain as fiscal consolidation continues, which raises the importance of boosting labor productivity and increasing the efficiency of the social protection. Introducing unemployment benefits may also help to improve matching in the labor market and provide support as the labor market softens. Other risks are related to the volatility in capital flows from the normalization of US monetary policy. The long-term sustainability of this favorable outlook hinges on structural reforms to strengthen medium-term fiscal planning, and to boost capabilities and competition within the economy.
Accelerated implementation of productivity-enhancing reforms to increase the quality of human capital and create more competition in the economy will be key for Malaysia to secure a lasting place among the ranks of high-income economies. Malaysia has been working to address these challenges. In 2010, Malaysia launched the New Economic Model (NEM), which aims for the country to reach high income status by 2020 while ensuring that growth is also sustainable and inclusive. The NEM includes a number of reforms to achieve economic growth that is primarily driven by the private sector and moves the Malaysian economy into higher value-added activities in both industry and services.
Heavily reliant on external trade, Malaysia’s economy has slowed for a second year because of weak demand for exports of hydrocarbons and manufactures and export prices that are mostly lower. Nevertheless, the gross domestic product is seen growing at a moderate pace this year, though slightly undershooting the forecast in March, with growth expected to pick up in 2017. Inflation in 2016 is lower than was anticipated and is projected to rise moderately in 2017. ADB retains the forecasts for the current account surplus to narrow in 2016 and expand again next year.
The Malaysian economy faces a challenging environment due to a continued plunge in commodity export prices, foreign exchange market turbulence – commodity-currency shocks, and a slowing Chinese economy. On the domestic front, the depreciation of the ringgit, increased federal government and household debts, and rising probability of default present a cloudy economic outlook.
The main problem is that such a developed country is dependent on external consequences. As world’s economy isn’t at its best place, Malaysia is experiencing a slowdown as well.
Different sources used to read modern scientific articles [7,8,9,10].
Robert Hill tell us about things happening in January 2016 in his article. He says that at its first policy meeting of 2016, which concluded on 21 January, Bank Negara Malaysia (BMN) decided to maintain the Overnight Policy Rate (OPR) at 3.25%, meeting markets’ expectations. This marks the ninth consecutive meeting in which BNM had left its policy rate unchanged, after having raised it to its current level in July 2014. However, the Bank did lower the Statutory Reserve Requirement Ratio, the percentage of eligible liabilities held in reserve accounts, from 4.00% to 3.50% percent. BNM has been forced to strike a balance between a steady outflow of capital, and an economy that is moderating. As a result, BNM opted to decrease the SRR in an effort to boost liquidity and enhance credit creation. The SRR has been at 4.00% since it was raised from 3.00% in July 2011. The lowering of the SRR is the latest in what the BNM calls its “comprehensive effort” to ensure appropriate levels of liquidity in the market. A cut in the SRR will free banks to lend more of the ringgit-denominated liabilities, which should have a positive effect on credit creation, and therefore on investment.
Joseph Chin continues with sudden decisions of July 2016. He reveals that Bank Negara Malaysia has unexpectedly reduced the Overnight Policy Rate (OPR) by 25 basis points to 3% at its Monetary Policy Committee (MPC) meeting, citing rising risks from Britain’s exit from the European Union. It said on Wednesday the ceiling and floor rates of the corridor for the OPR are correspondingly reduced to 3.25% and 2.75% respectively. This move could see banks lowering their lending rates and making it cheaper for eligible consumers and companies to take loans. Correspondingly, the saving rates could also go down. The decision gave the stock market a boost, especially property counters which had been languishing of late. The FBM KLCI closed up 6.42 points or 0.39% to 1,660.39 after the announcement was made at 3pm. Leading the list of gainers was UOA Development REIT, which jumped 30 sen to RM2.49 with 7.50 million shares done. In its statement, BNM said: “Looking ahead, there are increasing signs of moderating growth momentum in the major economies. Global growth prospects have also become more susceptible to increased downside risks in light of possible repercussions from the EU referendum in the United Kingdom. “International financial markets could also be subject to greater volatility going forward. In this light, global monetary conditions are expected to remain highly accommodative,” it said. Standard Chartered Global Research said the move to lower the OPR was “against our (and market) expectations”. “We have been making a non-consensus call for a rate cut this year since Q4-2015 given deteriorating fundamentals. However, the last MPC (in May) was neutral. This goes against the typical process, where the central bank communicates a potential change in stance before it sanctions a change,” it said.
Luis Lopez Vivas updates us upon September 2016 in Malaysia. On 7 September, Bank Negara Malaysia (BNM) held the fifth of its six monetary policy meetings this year. This was the third such meeting under the remit of newly-appointed governor Datuk Muhammad bin Ibrahim. After a surprising 25-basis-point cut in July, BNM maintained its Overnight Policy rate (OPR) unchanged at 3.00%—a decision widely expected by market analysts. The statutory reserve requirement (SRR) was also left unchanged at 3.50%. Although BNM’s decision follows Q2’s weak GDP figures—resulting from weaker net exports and a drawdown in stocks—the Bank is waiting to see if the government includes additional stimulus measures in its upcoming budget in October. This fiscal stimulus could potentially boost private consumption and private investment which were the key drivers of growth in Q2. On the external front, Bank Negara expects growth to remain subdued given the lackluster demand from Malaysia’s key trading partners and, overall, monetary authorities estimate the economy to grow within expectations in 2016, and to continue on a steady growth path next year. The Central Bank highlighted that domestic financial conditions had remained stable since July’s MPC meeting and that financial markets continued to operate in an organized manner. They also emphasized that, “at the current level of the OPR, the degree of monetary accommodativeness is consistent with the policy stance to ensure that the domestic economy continues on a steady growth path amid stable inflation, supported by continued healthy financial intermediation in the economy.”
- Statistical data and its description
All the data was collected from a trusted internet source such as DataBank of WorldBank.I collected information from 1997 up to 2015 which discloses affection of monetary policy on different indicators and external and internal factors that might of this indicators.
- Variables
Endogenous factors (dependent variables) are indicators reflecting the financial stability and economic growth of Malaysia (ex. GDP growth, GDP per capita etc.) while exogenous factors (independent variables) are ratios connected with the currency policy of Malaysia.
Endogenous variables:
- GDP growth, annual %
- GDP per capita, current US$
- Cash surplus/deficit, % of GDP
- Central government debt, total % of GDP
- Inflation consumer prices, annual %
- Current account balance, % of GDP
Exogenous variables:
- Foreign direct investment, net, BoP, current bln US$
- Inflation, consumer prices, annual %
- Net domestic credit, current bln LCU
- Total gold reserves, current US$
- Exchange rate, MYR to USD
- Average annual OPEC crude oil price, in U.S. dollars per barrel
- Description of statistical data related to the data
As I have mentioned, there are exogeneous and endogeneous variables. GDP growth is an annual percentage growth rate of GDP at market prices based on constant local currency. Aggregates are based on constant 2010 U.S. dollars. GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. GDP per capita is gross domestic product divided by midyear population. GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in current U.S. dollars. Current account balance is the sum of net exports of goods and services, net primary income, and net secondary income. Debt is the entire stock of direct government fixed-term contractual obligations to others outstanding on a particular date. It includes domestic and foreign liabilities such as currency and money deposits, securities other than shares, and loans. It is the gross amount of government liabilities reduced by the amount of equity and financial derivatives held by the government. Because debt is a stock rather than a flow, it is measured as of a given date, usually the last day of the fiscal year. Inflation as measured by the consumer price index reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. The Laspeyres formula is generally used. Net domestic credit is the sum of net claims on the central government and claims on other sectors of the domestic economy (IFS line 32). Data are in current local currency. Official exchange rate refers to the exchange rate determined by national authorities or to the rate determined in the legally sanctioned exchange market. It is calculated as an annual average based on monthly averages (local currency units relative to the U.S. dollar). All the statistical data are represented in Appendix.
- Construction of the econometric model
- Model specification
On the basis of macroeconomic theory all our models should be completed according to 4 principles of specification. The laws, which will be used, are translated into mathematical language. The number of equations are equal to the amount of endogeneous variables. All the indicators are dated. And, also, there is an influence of random disturbance. All these principles are used in the specification of my model
where:
Y_{1} — GDP growth, annual % | X_{1} — Foreign direct investment, net, BoP, current bln US$ |
Y_{2} — GDP per capita, current US$ | X_{2} — Inflation, consumer prices, annual % |
Y_{3} — Cash surplus/deficit, % of GDP | X_{3} — Net domestic credit, current bln LCU |
Y_{4} — Central government debt, total % of GDP | X_{4} — Total gold reserves, current US$ |
Y_{5} — Inflation, consumer prices, annual % | X_{5} — Exchange rate, MYR to US$ |
Y_{6} — Current account balance, % of GDP | X_{6} — Average annual OPEC crude oil price, in U.S. dollars per barrel |
- Steps of estimating the coefficients of the model
We have to estimate necessary coefficients. For this aim we will use MS Excel. We will do the linear regression analysis, which will demonstrate us the general form of our models through the calculation and analysis of coefficients, standard deviation and disturbance terms. To obtain these results we will use regression analysis. So, we turn to Data Analysis and choose function РЕГРЕССИЯ. All the calculations are shown in appendix.
- Specification of the estimated model
- Interpretation of the coefficients of the model
All coefficients were calculated in order to understand how different endogenous variables affect exogenous variables. Each coefficient corresponds to an exogenous variable:A1 corresponds to FDI; A2 corresponds to Inflation; A3 corresponds to net domestic credit; A4 corresponds to total gold reserves of Malaysia; A5 corresponds to exchange rate of MYR and USD; A6 corresponds to average annual OPEC crude oil prices in USD.
Let’s look at the example of explanation of the first model. Y1 corresponds to GDP growth. If FDI will increase by 1 bln USD, GDP growth will decrease by 0.65%. If inflation will increase by 1%, GDP growth will decrease by 1.38%. If net domestic credit will increase by 1 bln MYR, GDP growth will decrease by 0.006%. If total gold reserves will increase by 1bln USD, GDP growth will decrease by 3.77%. If exchange rate will increase by 1 unit, GDP growth will increase by 1.31%. If average annual OPEC crude oil price will increase by 1USD, GDP growth will increase by 0.2%.
So we can see that a positive coefficient illustrates positive relation between an endogenous variable and an exogenous one. Using this concept we can analyze and describe all the models.
- Tests: R^{2}-test, F-test, t-test, GQ-test, DW-test
R^2 Test
The coefficient of determination, named R^{2} or r^{2} , is a number that indicates how well data fit a statistical model. It provides a measure of how well observed outcomes are replicated by the model, as the proportion of total variation of outcomes explained by the model.
If the normalized R2 is close to 1, out regression model explains the values of Y with the given X values. We can find the number of the normalized R^2 in the regression analysis of a model. In the work, the value of normalized R^2 is higher than 0.5 in every model, which shows that more than 50% of total deviation of Yi is explained by the variation of the factors xi. That means that selected external factors significantly affect our endogenous variables, which confirms the correctness of their inclusion in the estimated model.
F-test
F-test is conducted in order to check the correctness of the R^2 value and the quality of the specification of the model. We must use data analysis to assess the needed variables. In order to calculate А critical we should use the formula (= F.ОБР.ПХ )[1]. If Fcrit is lower that F, than the quality of specification of our model is high. In the work Fcrit is lower than F in 4 models out of 5, so we can say that R^2 is non-random in those model observed (see Appendix). This test is not done successfully in the GDP growth model and it can be explained by the fact that rate itself as a mathematical index is rather independent as it compares a one time period data with its previous time period. This model would not be taken into consideration furthermore.
T-test
T-test checks the significance of coefficients F-test. In order to calculate t critical we should use the following formula: =СТЬЮДЕНТ.ОБР.2Х. T [1] of the coefficients of the model should be bigger than t_{critical }by modul. According to this test,net domestic credit, total gold reserves and OPEC oil price affect GDP per capita, FDI, inflation and total gold reserves affect cash surplus deficit, FDI, inflation and net domestic credit affect government debt and FDI, total gold reserves and exchange rate to USD affect current account balance.
GQ-test
This test is designed to check the conditions of the Gauss-Markov theorem. Goldfeld–Quandt test checks for homoscedasticity in regression analyses. If residuals turn out to be homoscedastic, then we may use ordinarily square in order to estimate parameters of the model. It does this by dividing a dataset into two parts or groups, and hence the test is sometimes called a two-group test. Steps of further calculations are described in the methodical book. [1] GQ and 1/GQ should be less than the newly calculated F critical and this condition is proved in every model of this work.
DW-test
It is a test statistic used to detect the presence of autocorrelation (a relationship between values separated from each other by a given time lag) in the residuals (prediction errors) from a regression analysis. Steps of calculations are described in the book [1] and results are calculations themselves are presented in the Appendix. If value of DW coefficient lies between 0 and dl or 4-dl and 4 than there is autocorrelation of residuals, if value of DW lies between du and 4-du than there is no autocorrelation in residuals, in other cases it is not certain. We can see from the results of this test that there is no information about autocorrelation in all models.
- Construction of confidence interval
We have to check confidence intervals in order to make conclusions about the adequacy of the model. We have to make sure that real value lies between lower and upper boundaries of Y. Calculations are presented in Appendix.
- Checking of model adequacy
According to results of constructing the confidence intervals we can see that cash surplus deficit and current account models are not adequate, while this condition is perfectly proved in GDP per capita and government debt models as all real values lie within calculated ones.
- Economic analysis of model results
From all the tests made we can understand what variables affect major macroeconomic indexes. We obtained that selected external factors significantly affect our endogenous variables in different ways in every model, which confirms the fact that currency policy affects the economy of Malaysia. We see that the most effective tool of the currency policy for Malaysia is the total gold reserves and in order to promote economy Malaysian government should focus on increasing it. Though despite government policies to increase income per capita in order to hasten the progress towards high income country by 2020, Malaysia growth in labour productivity and wages has been very slow, lagging behind by the OECD standard.
- Model forecasting
Using coefficients from regression analysis and forecasting exogenous variables for the next year by finding tendency looking at the diagrams we calculated the forecasted endogenous variables. We see that in the adequate models (GDP per capita and government debt) forecasts are opposite. as forecast for GDP per capita is optimistic predicting it will rise what means that currency policy affects macroeconomic indicators in a positive way and forecast for government debt is pessimistic as it is predicted to rise too what can’t be said to be good for an economy.
- Conclusions
We observed statistical data of Malaysia, which is a member of ASEAN, and did research about dependence of major macroeconomic indicators, such as GDP growth rate, GDP per capita, cash surplus deficit as % of GDP, government debt as % of GDP and current account balance, from independent variables, including FDI, inflation, net domestic credit, total gold reserves, exchange rate with USD and OPEC oil prices. Our calculations showed that some models were constructed wisely, dependence exists and exogenous variables really affect country’s economy, whilst some have no interdependence, which can be explained by a number of factors.
- Recommendations
We can see that in order to increase GDP per capita Malaysian government should focus on increasing its gold reserves, which play a significant role in forming a currency policy of a country. We can also observe that both GDP per capita and central government debt are highly dependent on the exchange with USD, which is formed from a very big number of implementations, so we can’t recommend to just lower the number of exchange as it is impossible.
- References
- Tregub I.V. Econometrical research. Methodological materials for the preparation for current control. –M.:Finance University, 2015.
- Tregub I.V. Forecasting economic indicators. — M.: PSTM, 2009.
- Tregub I.V. Mathematical models of the dynamics of economic systems. — M.: Finance Academy, 2009.
- Tregub I.V. Features of investment in the innovative projects / I.V. Tregub // Economics.Taxes.Law. — 2013.
- Website: [http://www.worldbank.org/]
- Website: [http://www.acean.org/]
- Website: [http://www.bloomberg.com/news/articles/2016-08-23/quicktake-q-a-oil-steers-path-of-malaysian-economy-currency]
- Website: [http://www.focus-economics.com/countries/malaysia/news/monetary-policy/central-bank-maintains-policy-rate-at-325-1]
- Website: [http://www.thestar.com.my/business/business-news/2016/07/13/bnm-cuts-opr-to-3pct-at-its-monetary-policy-committee/]
- Website: [http://www.focus-economics.com/countries/malaysia/news/monetary-policy/bnm-maintains-its-policy-rate-unchanged-at-300-after-julys]
Appendix.
Endogenous variables (statistical data)
Year | GDP growth, annual % | GDP per capita, current US$ | Cash surplus/deficit, % of GDP | Central government debt, total % of GDP | Inflation, consumer prices, annual % | Current account balance, % of GDP |
Y_{1} | Y_{2} | Y_{3} | Y_{4} | Y_{5} | Y_{6} | |
1997 | 7,32 | 4585,69 | 2,92 | 31,80 | 2,66 | -5,93 |
1998 | -7,36 | 3227,81 | -0,82 | 36,10 | 5,27 | 13,20 |
1999 | 6,14 | 3456,50 | -3,83 | 36,90 | 2,74 | 15,92 |
2000 | 8,86 | 4004,56 | -4,12 | 35,30 | 1,53 | 9,05 |
2001 | 0,52 | 3878,77 | -3,51 | 41,40 | 1,42 | 7,85 |
2002 | 5,39 | 4132,67 | -4,90 | 43,10 | 1,81 | 7,13 |
2003 | 5,79 | 4431,24 | -4,85 | 45,10 | 0,99 | 12,14 |
2004 | 6,78 | 4924,59 | -4,12 | 45,70 | 1,52 | 12,09 |
2005 | 5,33 | 5564,17 | -3,76 | 44,40 | 2,96 | 13,92 |
2006 | 5,58 | 6194,67 | -2,96 | 43,20 | 3,61 | 16,10 |
2007 | 6,30 | 7240,68 | -3,17 | 42,70 | 2,03 | 15,38 |
2008 | 4,83 | 8486,60 | -4,41 | 42,80 | 5,44 | 16,86 |
2009 | -1,51 | 7312,01 | -6,13 | 55,40 | 0,58 | 15,72 |
2010 | 7,43 | 9069,04 | -5,00 | 53,10 | 1,71 | 10,06 |
2011 | 5,29 | 10427,76 | -4,62 | 51,80 | 3,20 | 10,90 |
2012 | 5,47 | 10834,66 | -4,39 | 53,30 | 1,66 | 5,19 |
2013 | 4,71 | 10973,66 | -4,36 | 54,70 | 2,11 | 3,47 |
2014 | 5,99 | 11307,06 | -4,13 | 52,70 | 3,14 | 4,39 |
2015 | 4,95 | 9766,17 | -3,97 | 53,50 | 2,10 | 3,02 |
Exogenous variables (statistical data)
Year | Foreign direct investment, net, BoP, current bln US$ | Inflation, consumer prices, annual % | Net domestic credit, current bln LCU | Total gold reserves, current bln US$ | Exchange rate, MYR to USD | Average annual OPEC crude oil price, in U.S. dollars per barrel |
X_{1} | X_{2} | X_{3} | X_{4} | X_{5} | X_{6} | |
1997 | -5,14 | 2,66 | 460,32 | 0,68 | 3,88 | 18,86 |
1998 | -2,16 | 5,27 | 459,21 | 0,68 | 3,80 | 12,28 |
1999 | -2,47 | 2,74 | 451,46 | 0,34 | 3,80 | 17,44 |
2000 | -1,76 | 1,53 | 493,16 | 0,32 | 3,80 | 27,60 |
2001 | -0,29 | 1,42 | 516,64 | 0,32 | 3,80 | 23,12 |
2002 | -1,30 | 1,81 | 550,45 | 0,40 | 3,80 | 24,36 |
2003 | -1,10 | 0,99 | 585,64 | 0,49 | 3,80 | 28,10 |
2004 | -2,56 | 1,52 | 604,33 | 0,51 | 3,80 | 36,05 |
2005 | -0,99 | 2,96 | 639,55 | 0,60 | 3,78 | 50,59 |
2006 | -0,05 | 3,61 | 683,77 | 0,74 | 3,53 | 61,00 |
2007 | 2,74 | 2,03 | 728,07 | 0,98 | 3,31 | 69,04 |
2008 | 7,83 | 5,44 | 853,48 | 1,02 | 3,45 | 94,10 |
2009 | 6,63 | 0,58 | 934,22 | 1,27 | 3,42 | 60,86 |
2010 | 4,46 | 1,71 | 1012,76 | 1,64 | 3,08 | 77,38 |
2011 | 2,99 | 3,20 | 1134,27 | 1,79 | 3,17 | 107,46 |
2012 | 8,00 | 1,66 | 1261,15 | 1,95 | 3,06 | 109,45 |
2013 | 2,11 | 2,11 | 1409,36 | 1,41 | 3,28 | 105,87 |
2014 | 5,44 | 3,14 | 1554,67 | 1,39 | 3,50 | 96,29 |
2015 | -1,08 | 2,10 | 1675,54 | 1,30 | 4,29 | 49,49 |
1. GDP growth model
- GDP per capita model
- Cash surplus deficit model
- Government debt model
- Current account balance model