Topic factors affecting gross private domestics investment of united state from 2001 to 2023

Econometrics Asignment NATIONAL ECONOMICS UNIVERSITY Faculty of Economics -  - ASSIGNMENT SUBJECT: ECONOMETRICS Topic: Factors affecting Gross Private Domestics Investment of United State from 2001 to 2023 Class: FE64 Group: 11 Instructor: Bui Duong Hai Ha Noi, 2023 Group members No Student ID Name Contribution Tỷ lệ % đóng góp 11223730 Phạm Thuỳ Linh 11225239 Nguyễn Hà Phương Research and make hypothetical models 35% 11224510 Nguyễn Quỳnh Nga Study and collect data 30% Estimate and analysis models 35% Scoring table Point - 10 7-8 5-6 3-4 0-2 Topic Section References Data Outcome Annalysis Presentation TABLE OF CONTENTS Introduction 1.1 Reason for choosing the topic 1.2 Objectives of the study 1.3 Research question 1.4 Object and scope of the study 2 Overview of theory and previous research 2.1 Theory 2.2 Some referenced researchs 2.3 General models Data analysis 3.1 Data 3.2 Estimated results 3.2.1 Selected models 3.2.2 Models comparison 3.3 Results meaning analysis 10 Summary 11 * Summarize the implementation process 11 * Tools 12 * Our capacity limitations 12 * Reference resources 12 APPENDIX 13 Econometrics Asignment Introduction 1.1 Reason for choosing the topic Gross Private Domestic Investment, or GPDI, is a measure of the amount of money that domestic businesses invest within their own country GPDI constitutes one component of GDP, which politicians and economists use to gauge a country’s overall economic activity The US is one of the world's largest and most diverse economies Thus, research on the factors that influence GPDI can help better understand how these factors interact in a diverse and volatile economic environment The period from Q1 2001 to Q2 2023 includes many important economic fluctuations such as the financial crisis in 2008 and the impact of the COVID-19 pandemic from 2019 to present Following the global financial crisis in 2008-2009, GPDI in the United States initially experienced a significant decline as businesses faced economic uncertainty However, in the subsequent years, there was a recovery as the economy stabilized Throughout much of the 2010s, the US experienced a period of economic expansion GPDI played a crucial role in this phase as businesses invested in capital goods, technology, and infrastructure Low-interest rates, along with increased consumer and business confidence, contributed to this growth The Tax Cuts and Jobs Act, enacted in December 2017, aims to stimulate economic growth by reducing corporate tax rates The impact of this tax reform on GPDI was evident as some businesses increased investment due to improved after-tax profitability In 2020, the global COVID-19 pandemic led to a significant economic downturn Many businesses reduced investment amid uncertainties, and the overall economic activity contracts The negative impact on GPDI was notable during the height of the pandemic In the latter part of 2020 and into 2021, the US began to recover from the pandemic's economic effects GPDI rebounded as businesses adapted to new conditions, and economic stimulus measures were implemented Therefore, the study can provide detailed insight into hypothetical different factors in these events influence GPDI Additionally, studying the factors that influenced GPDI during this period could provide useful information to understand current trends and challenges facing the United States in attracting and retaining domestic investment 1.2 Objectives of the study The main objective of the topic is to identify and analyze factors and analyze the impact of those factors on Gross Private Domestics Investment of United State from first quarter 2001 to second quarter 2023 To achieve the above general research goal, the research aims at the following specific goals: Firstly, clarify the theoretical basis of factors affecting Gross Private Domestics Investment of United State Secondly, identify, analyze and evaluate the influence of factors affecting Gross Private Domestics Investment of United State from first quarter 2001 to second quarter 2023 1.3 Research question Questions identified include: Firstly, what theoretical content is related to the Gross Private Domestics Investment of United State? FE64 - Group 11 Econometrics Asignment Secondly, are there any special interactions between these factors and Gross Private Domestics Investment of United State? 1.4 Object and scope of the study − Research object: Factors affecting Gross Private Domestics Investment of United State from first quarter 2001 to second quarter 2023 − Research scope: spatial scope: United State; time scope: first quarter 2001 to second quarter 2023 The research team decided to research over 20 years to cover many different economic periods, from the post-dot-com bubble to the great recession of 2008 and recovery This provides a rich and varied view of how factors influencing GPDI may change and interact over time − Data and data sources: Research data was compiled by the authors from the first quarter of 2001 to the second quarter of 2023 with 90 observations taken from the following websites: • • • • • • Gross Private Domestic Investment (GPDI) | FRED | St Louis Fed (stlouisfed.org) Rest of the World; Foreign Direct Investment in U.S.; Asset (Current Cost), Level (ROWFDNQ027S) | FRED | St Louis Fed (stlouisfed.org) Interest Rates and Price Indexes; Effective Federal Funds Rate (Percent), Level (BOGZ1FL072052006Q) | FRED | St Louis Fed (stlouisfed.org) Federal government current tax receipts (W006RC1Q027SBEA) | FRED | St Louis Fed (stlouisfed.org) Unemployment Rate: Aged 15-64: All Persons for United States (LRUN64TTUSQ156S) | FRED | St Louis Fed (stlouisfed.org) Recession: https://www.investopedia.com/articles/economics/08/past-recessions.asp Overview of theory and previous research 2.1 Theory - Theory related to research question The Gross Private Domestic Investment (GPDI) is a key economic indicator reflecting the total value of capital expenditures made by businesses within a country's borders Foreign Direct Investment (FDI) plays a significant role in influencing GPDI, as it represents capital flows from abroad contributing to domestic investment A higher influx of FDI can stimulate economic growth, bolstering GPDI through increased business activities and capital expenditures Conversely, domestic factors such as taxation, unemployment, and interest rates also wield substantial influence Tax policies directly impact the after-tax returns on investments, shaping businesses' decisions to allocate resources toward capital projects Unemployment rates can impact consumer spending, affecting demand for goods and services, thereby influencing businesses' investment decisions Interest rates play a dual role by affecting the cost of borrowing for investment projects and influencing the opportunity cost of capital, both of which can impact GPDI A holistic understanding of FDI, tax policies, unemployment, and interest rates is essential for policymakers and businesses alike, as they collectively shape the landscape in which private domestic investments unfold - Theory (briefly) about factors FE64 - Group 11 Econometrics Asignment Factors affecting Gross Private Domestics Investment Firstly, impact of tax on investment behavior In accordance with the Neoclassical Investment Model, the impact of taxes on investment behavior is intricately tied to the after-tax cost of capital Within this theoretical framework, higher tax rates are perceived as elevating the cost of capital for businesses As taxes take a larger share of a firm's income, the post-tax income available for reinvestment diminishes This shift in the cost structure can result in a diminished overall attractiveness of potential investment projects, as projects that once met profitability thresholds may now incur higher after-tax costs Second, impact of unemployment rate on investment behavior The impact of unemployment on investment behavior is intricately linked to the concept of aggregate demand in economic theory High levels of unemployment can exert downward pressure on aggregate demand as individuals facing joblessness experience reduced incomes and, consequently, decreased purchasing power A decline in aggregate demand can lead to diminished sales prospects for businesses so companies may reassess and potentially scale back their investment plans As businesses anticipate weaker market conditions, they may delay capital expenditures and new projects, aligning their investment decisions with the subdued economic environment Third, impact of foreign direct investment on investment behavior Inflow of capital from abroad in the form of private investment is important for the growth of developing the economy; especially at the initial stage of its economic development Foreign investment brings various positive impacts, playing a pivotal role in fostering technological advancements, enhancing managerial expertise, boosting export capabilities, and contributing to overall economic growth Forth, impact of interest rate on investment behavior Due to the opportunity cost, high-interest rates may make alternative, less risky investments more attractive than business investments Investors may choose to allocate their funds to interest-bearing assets rather than engaging in capital-intensive projects 2.2 Some referenced researchs Gashi B, Asllani Gani, Boqolli L (2018) The effect of tax structure in economic growth The main goal of this paper is to analyse the effect of the tax structure in the economic growth of Kosovo in the period 2007-2015 The study intends to evaluate the impact of specific types of taxes on economic growth The econometric model includes several independent variables (types of taxes), and the dependent variable GDP Based on data obtained through the log-log model, the results show the impact of special taxes such as Pt, It, VAT, Wt, Ibt, Tdr , Ct on GDP The results show that most of the taxes have a positive impact on GDP growth; it is also shown that not all taxes have the same impact on economic growth In the econometric analysis the coefficient of R2=0,999 reflects the high degree of determination with 99.9% forecasting accuracy Donny Susilo (2018) The Impact of Foreign Direct Investment on Economic Growth (a Causal Study in the United States) This research examines the impact of Foreign Direct Investment on Economic Growth in the United States by multiple linear regression model and its estimation using ordinary least squares (OLS) This research classifies all the sectors into FE64 - Group 11 Econometrics Asignment 10 sectors This research uses data for the period 2000 –2017 and suggests that not all forms of foreign investment seem to be beneficial to host economies Some sectors provide positive correlation to economic growth and some provide negative effects Nevertheless, it is significant yet, this is because there are different characteristics between developed and developing countries Economic growth in the U.S is mostly driven by personal consumption Shrikant Panigrahi, Noor Azizan, Shahryar Sorooshian, Prabha Thoudam (2020) Effects of Inflation, Interest, and Unemployment Rates on Economic Growth: Evidence from ASEAN Countries This study using secondary panel data gathered from the World Bank database to investigate the long-run relationship between these factors and GDP growth from 1995 to 2018 in ASEAN-5 countries (Malaysia, Indonesia, Thailand, Singapore and the Philippines) Statistical results show a strong dynamic long-run linkage between interest and inflation rates and economic growth, but the linkage between unemployment rate and economic growth is insignificant Granger’s test of causality indicates that interest, unemployment and inflation rates and economic growth are related Policy makers should be aware of these relationships when making decisions to facilitate economic growth and stability 2.3 General models Population Regression Model 𝐺𝑃𝐷𝐼 = 𝛽1 + 𝛽2 𝐹𝐷𝐼 + 𝛽3 𝑇𝐴𝑋 + 𝛽4 𝐼 + 𝛽5 𝑈 + 𝛽6 𝑅𝐸𝐶𝐸𝑆𝑆 + 𝑒 Sample Regression Model ̂1 + 𝛽 ̂2 𝐹𝐷𝐼 + 𝛽 ̂3 𝑇𝐴𝑋 + 𝛽 ̂4 𝐼 + 𝛽 ̂5 𝑈 + 𝛽 ̂6 𝑅𝐸𝐶𝐸𝑆𝑆 ̂ =𝛽 𝐺𝑃𝐷𝐼 Table 1: The variables in researched models Hypothesis Variables Code Unit Dependent variable Gross Private Domestic Investment GPDI Billions of Dollars FDI TAX Millions of Dollars Billions of Dollars + + I U RECESS Percentage Percentage None + - Independent variables H1 Foreign Direct Investment H2 Tax receipts of Federal Government H3 Interest Rate H4 Unemployment Rate H5 Recession periods Expected sign Within this set, the recession variable signifies times when the US economy undergoes a crisis Include - The Dot-Bomb Recession: Q1 2001 – Q4 2001 - The Great Recession: Q1 2007 – Q2 2010 - The COVID-19 Recession: Q1 2020 – Q4 2022 FE64 - Group 11 Econometrics Asignment Data analysis 3.1 Data Table 2: Descriptive Statistics GPDI FDI TAX I U RECESS Mean 2945.379 3374873 1735.512 1.503889 5.975212 0.333333 Median 2704.235 3036869 1604.644 1.005000 5.484200 0.000000 Maximum 4796.162 6382525 3131.082 5.310000 13.27682 1.000000 Minimum 1841.416 1517843 994.7080 0.070000 3.580972 0.000000 Std Dev 832.3463 1434442 537.7937 1.671904 2.000255 0.474045 Skewness 0.642714 0.481774 0.794374 1.078335 1.086995 0.707107 Kurtosis 2.478816 2.022431 3.153432 2.914121 3.804440 1.500000 Jarque-Bera 7.214853 7.065249 9.553728 17.46975 20.15009 15.93750 Probability 0.027122 0.029228 0.008422 0.000161 0.000042 0.000346 Sum 265084.1 3.04E+08 156196.1 135.3500 537.7691 30.00000 Sum Sq Dev 61659230 1.83E+14 25740763 248.7783 356.0907 20.00000 Observations 90 90 90 90 90 90 From this table, we can use the probability of Jarque-Bera in order to test for normality distribution of variables Assume: H0: The variable distributes in normality H1: The variable does not distribute in normality If p-value (Jarque-Bera test) < 5% → Reject H0: The variable does not distribute in normality Otherwise, the variable distributes in normality It is witnessed that all variables are not normally distributed Table 3: Covariance and Correlation coefficients Covariance Correlation Probability GPDI 685102.6 FDI TAX I U RECESS GPDI 1.000000 - FDI TAX I U RECESS FE64 - Group 11 1.12E+09 2.03E+12 0.950729 1.000000 0.0000 - 431481.4 7.17E+08 286008.5 0.974754 0.940206 1.000000 0.0000 0.0000 - 21.09654 -444093.2 -25.45207 2.764204 0.015330 -0.187255 -0.028625 1.000000 0.8860 0.0772 0.7888 - -806.7569 -759135.1 -449.7584 -1.842383 3.956563 -0.490011 -0.267549 -0.422796 -0.557103 1.000000 0.0000 0.0108 0.0000 0.0000 - 30.79001 92385.46 42.15085 0.062370 0.099877 0.222222 0.078911 0.137389 0.167195 0.079579 0.106516 1.000000 0.4597 0.1966 0.1152 0.4559 0.3177 - Econometrics Asignment Regarding of covariance and correlation coefficients, it is clear that there is a positive relationship among GPDI and other variables, which is the same as expected with the exception of unemployment rate 3.2 Estimated results 3.2.1 Selected models − General model Population regression model: 𝐺𝑃𝐷𝐼 = 𝛽0 + 𝛽1 𝐹𝐷𝐼 + 𝛽2 𝑇𝐴𝑋 + 𝛽3 𝐼 + 𝛽4 𝑈 + 𝛽5 𝑅𝐸𝐶𝐸𝑆𝑆 + 𝑒 Sample regression model: ̂0 + 𝛽 ̂1 𝐹𝐷𝐼 + 𝛽 ̂2 𝑇𝐴𝑋 + 𝛽 ̂3 𝐼 + 𝛽 ̂4 𝑈 + 𝛽 ̂5 𝑅𝐸𝐶𝐸𝑆𝑆 ̂ =𝛽 𝐺𝑃𝐷𝐼 Table 4: Estimated results Dep: C FDI GPDI [1] [2] 1001.735 1168.543 *** *** 0.0003 0.0003 *** *** LOG(GPDI) [3] 375.732 *** 0.0002 *** LOG(FDI) TAX 0.717 *** 0.743 *** U 30.457 *** -49.589 *** 24.454 *** -65.684 *** 1/U RECESS -105.023 *** -84.721 *** 1801.312 *** -118.059 *** 0.000 0.984 0.983 103.589 2.864 0.487 0.961 0.701 0.000 0.041 0.000 0.982 0.982 109.842 3.193 0.439 0.806 0.602 0.000 0.152 0.000 0.982 0.981 109.828 2.987 0.5499 0.617 0.2298 0.000 0.004 U*RECESS F-stat (pv) R-sq Adj R-sq RMSE MAPE DW Ramsey (pv) White (pv) BG (pv) JB (pv) FE64 - Group 11 [5] -0.334 [6] 1.14979 *** [7] 0.862 *** [8] 0.913 *** 0.241 *** 0.568 *** 0.149 *** 0.215 *** 0.165 *** 0.642 *** 0.544 *** 0.598 *** -0.0298 *** -0.023 0.809 *** LOG(TAX) I [4] 0.624 ** 0.519 *** 0.007 * -0.0204 *** -0.039 *** 0.016 *** -0.035 *** -0.026 * *** 0.839 *** -0.042 *** -0.1498 -0.035 *** *** 0.019 *** 0.000 0.000 0.000 0.000 0.000 0.971 0.947 0.974 0.97 0.969 0.969 0.945 0.972 0.969 0.968 129.489 193.621 126.428 132.37 134.019 3.314 5.235 3.368 3.452 3.337 0.373 0.243 0.512 0.397 0.409 0.268 0.151 0.251 0.34 0.617 0.0598 0.001 0.187 0.015 0.099 0.000 0.000 0.000 0.000 0.000 0.026 0.699 0.554 0.049 0.014 Econometrics Asignment With significant level: 0.01***, 0.05**, 0.1* It is clear that all models have significant F-test, by that at least one variable can explain for its overall model Looking at the Adjusted R square of eight models, these values are very high The highest belongs to model [1], which means that 98.3% of the dependent variable GPDI is explained by the model 3.2.2 Models comparison Table 5: Models comparision DW Ramsey RESET White heter BG test JB R2 RMSE MAPE [1] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Nonnormally distributed random errors 0.984 103.589 2.864 [2] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Normally distributed random errors 0.982 109.842 3.193 [3] Auto correlation order No omitted variable Homoske dasticity Auto correlation order 0.982 109.828 2.987 [4] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Nonnormally distributed random errors Nonnormally distributed random errors 0.971 129.489 3.314 [5] Auto correlation order No omitted variable Heterosk edasticit y Auto correlation order Normally distributed random errors 0.947 193.621 5.235 [6] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Normally distributed random errors 0.974 126.428 3.368 [7] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Nonnormally distributed random errors 0.969 133.483 3.3197 FE64 - Group 11 Econometrics Asignment [8] Auto correlation order No omitted variable Homoske dasticity Auto correlation order Nonnormally distributed random errors 0.969 134.019 3.337 The bolded part illustrates the model does not violate the hypotheses In terms of this comparison, it is witnessed that all models have autocorrelation orders and Model [5] exists in Heteroskedasticity, while models [1] [3] [4] [7] [8] not have normal distribution of errors Fortunately, there is no error in the functional form of any model In conclusion, model [2] and [6] can be considered as the models with the best estimated results among eight models Nevertheless, we chose model [6] The reason for this choice is simply because we expected the relationship among variables is not linearity The relationship between dependent variable and independent variables resemble to expected sign - Chosen model results: Population regression: 𝐿𝑂𝐺(𝐺𝑃𝐷𝐼) = 𝛽0 + 𝛽1 𝐿𝑂𝐺(𝐹𝐷𝐼) + 𝛽2 𝐿𝑂𝐺(𝑇𝐴𝑋) + 𝛽3 𝑈 + 𝛽4 𝑅𝐸𝐶𝐸𝑆𝑆 + 𝛽5 𝑈 ∗ 𝑅𝐸𝐶𝐸𝑆𝑆 + 𝑒 Sample regression: ̂ ̂0 + 𝛽 ̂1 𝐿𝑂𝐺(𝐹𝐷𝐼) + 𝛽 ̂2 𝐿𝑂𝐺(𝑇𝐴𝑋) + 𝛽 ̂3 𝑈 + 𝛽 ̂4 𝑅𝐸𝐶𝐸𝑆𝑆 + 𝛽 ̂5 𝑈 ∗ 𝑅𝐸𝐶𝐸𝑆𝑆 𝐿𝑂𝐺(𝐺𝑃𝐷𝐼) =𝛽 This model shows that: − The regression function is suitable because the F test has P–value = 0.000< α = 0.05 − 97.40% of the sample variation in the dependent variable GPDI in the sample is explained by the model Estimated meaning of the coefficients: 𝛽0 = 1.15: when all independent variables have no effect on GPDI, the average estimate of 𝐿𝑂𝐺(𝐺𝑃𝐷𝐼) is 1.15, by that, approximate 3.16 billion Dollars 𝛽1 = 0.149: when 𝐹𝐷𝐼 increases by 1%, ceteris paribus, on average, estimated change of 𝐺𝑃𝐷𝐼 is 0.149% 𝛽2 = 0.642: when TAX increases by 1%, ceteris paribus, on average, estimated change of GPDI is 0.642% 𝛽3 = -0.0298: when 𝑈 increases by unit, ceteris paribus, on average, estimated change of GPDI increases (e-0.0298-1) *100%, decreases approximately –2.94% 𝛽4 = -0.1498: in recession periods (𝑅𝐸𝐶𝐸𝑆𝑆 = 1), when all independent variables have no effect on GPDI, the average estimate of LOG(GPDI) is lower by 0.1498 𝛽5 = 0.019: in recession periods (𝑅𝐸𝐶𝐸𝑆𝑆 = 1), when U increase by unit, ceteris paribus, on average, estimated change of 𝐿𝑂𝐺(𝐺𝑃𝐷𝐼) higher by 0.019 ≫ Check for the suitability of the model Pair of hypotheses: H0: The regression model is not appropriate H1: The regression model is suitable From the EViews results, we see that P-value = 0.000 < 0.05 → Reject H0, with a significance level of 5%, the regression model is appropriated FE64 - Group 11 Econometrics Asignment ≫ Check for perfect multicollinearity From Table 3, the correlation between dependent variable (GPDI) and independent variables (FDI and TAX) is very high (0.951 and 0.975 respectively), the same as we expected Nevertheless, the correlation among independent variables should not be expected highly If the correlation is greater than 0.8, there may exist high multicollinearity For example, in our report, the correlation between FDI and TAX is 0.94 (it may exist high multicollinearity) Table 6: Variance Inflation Factors Coefficient Uncentered Centered Variable Variance VIF VIF C 0.044565 1897.922 NA LOG(FDI) 0.001749 16637.40 13.94594 LOG(TAX) 0.004242 9946.231 16.07134 U 1.45E-05 24.41514 2.435722 RECESS 0.001299 18.44018 12.29345 U*RECESS 3.09E-05 19.96372 14.20291 Centered VIF of LOG(FDI) and LOG(TAX) are very high (>10), there may exist high multicollinearity between these two variables However, the p-value (t-stat) in all models implies that LOG(TAX) has high meaning in the contribution of overall model Therefore, we decided to keep both LOG(FDI) and LOG(TAX) in our models ≫ Ramsey test for error functional form Test hypothesis: H0: The original model no error functional form H1: The original model has omitted variable Inspection results: Ramey RESET test for fitted term t-statistic F-statistic Likelihood ratio Value 1.156373 1.337199 1.438418 df 83 (1, 83) Probability 0.2508 0.2508 0.2304 According to this result, P–value (F-test) = 0.2508 is greater than the significance level of 0.05, which is not enough basis to reject H0, the original model has the correct functional form and there are unbiased estimator coefficients ≫ White test for heteroskedasticity (no cross term) Test hypothesis: H0: The original model has homoskedasticity H1: The original model has heteroskedasticity Inspection results: White Heteroskedasticity test for no cross term F-statistic FE64 - Group 11 1.536812 Prob F(5,84) 0.1872 Econometrics Asignment Obs*R-squared Scaled explained SS 7.542921 7.329670 Prob Chi-Square(5) Prob Chi-Square(5) 0.1833 0.1973 According to this result, P-value (F-test) = 0.1872 is greater than the significance level of 0.05, which is not enough basis to reject H0, the original model does not have heteroscedasticity and there are efficient estimator coefficients ≫ Check for autocorrelation - Durbin Watson test for first order serial correlation: DW=0.512 → 𝜌 ≈ 0.774, there is a positive serial correlation order - Breusch-Godfrey test for second order serial correlation: Test hypothesis: H0: The model has no serial correlation order H1: The model has serial correlation order Inspection results: Breusch-Godfrey Serial Correlation LM Test: F-statistic Obs*R-squared 51.43135 50.07848 Prob F(2,82) Prob Chi-Square(2) 0.0000 0.0000 According to this result, P-value (F-test) = 0.000 is less than the significance level of 0.05 → Reject H0, the model has serial correlation order ≫ Jarque-Bera test for normally distribution of error Test hypothesis: H0: The model has normally distribution of error H1: The model has no normally distribution of error Inspection results: Mean Median Maximum Minimum Std Dev Skewness Kurtosis JarqueBera Probabilit y -1.44E-15 -0.002804 0.122991 -0.101146 0.044661 0.255759 3.231009 1.181307 0.553965 According to this result, P-value (F-test) = 0.554 is greater than the significance level of 0.05, which is not enough basis to reject H0, the original model has normally distribution of error ≫ Compare to the expectations From the estimated results, it is clear that when FDI and government tax increase lead to an increase in GPDI Besides, when unemployment rate rises or the economy is in recession periods, which will lead to a drop of GPDI All are similar to expectation, 3.3 Results meaning analysis Questions answer: - Firstly, what theoretical content is related to the Gross Private Domestics Investment of United State? From the above regression analysis results, it shows that all models are appropriate, however models containing interest rates still violates the White test, so we removed variable I in FE64 - Group 11 10 Econometrics Asignment some models So there are important relationships between FDI and some factors, such as Foreign Direct Investment, government tax receipts and unemployment rate It is suspected that Diminishing Return to Scale will exist when FDI and taxes increase to a certain level, so the log-log functional form is preferred - Secondly, are there any special interactions between these factors and Gross Private Domestics Investment of United State? Comparing models [6] and [7], model [6] include the interaction of unemployment rate and recessions, there are unbiased and almost efficient of estimator coefficients The results are quite surprising, it can be seen that during the economic crisis, when there was no impact of the independent variables and when there was a change in the unemployment rate it had a greater impact, specifically on employment GPDI is higher than when there was no financial crisis, thereby slightly reducing the impact of FDI on GPDI Possible causes are: - - - Stimulus Packages: Many countries may deploy economic stimulus packages to support businesses and people These measures could include tax breaks, or providing direct financial support These measures can stimulate investment and production, thereby increasing GDPDI It can also be seen from the data that taxes has larger effect on GPDI than other variables, also the correlation of FDI and tax repceipts is quite high Stock market: In some cases, the stock market can reflect strange growth amid a crisis This may be due to investors' expectations of economic recovery, leading to investments in companies with growth potential Flexible monetary policy: If the central bank implements flexible monetary policy and reduces interest rates, this can stimulate spending and investment, helping to lift economic activity and increase GDP It can be seen that the US has had a number of policies to limit the heavy decline in domestic investment during periods of financial crisis Summary The author group chose the topic "Factors affecting Gross Private Domestics Investment of United State from the first quarter of 2001 to second quarter of 2023" Collecting data from 90 quarters in the field of United State’s macroeconomics on the Federal Reserve Economic Data website, including variables: GPDI, FDI, TAX, U, RECESS Then use EVIEW 10 software to see the correlation between variables, estimate and select the best model for the topic * Summarize the implementation process - Give a hypothetical model - Researching and studying US newspapers - Making questions and hypothesis - Collecting data and expect FE64 - Group 11 11 Econometrics Asignment - Running estimation and statistics - Analyzing data and compare with initial expectations - Summarizing results and conclusions - Making report and self-assessment * Tools Using “EViews 10” software to estimate, “Excel” and “Word” on my.sharepoint.com to complete the assignment * Our capacity limitations -Our assignment may exist some mistake because of our skill limitations -There are some violations in OLS assumptions and we tried to fix it by changing the functional forms, removing or adding variables, re-estimating the standard error -The data does not completely efficient * Reference resources We have information collected and data from economic statistics websites in US and have chosen the website https://fred.stlouisfed.org/ to get the most accurate data - Printed documents [1] Wooldridge J.M (2016), Introductory Econometrics A modern Approach, 6th Editon, Cengage Learning - Online documents [3] U.S Bureau of Economic Analysis, Gross Private Domestic Investment [GPDI] (From Q1/2001 to Q2/2023) retrieved from FRED, Federal Reserve Bank of St Louis; https://fred.stlouisfed.org/series/ ( access date : 13/10/2023) [4] Pratibha S.Gaikwad, Gholamreza Fathipour (2013) The Impact of Foreign Direct Investment (FDI) on Gross Domestic Production (GDP) in Indian Economy, Information Management and Business Review Vol 5, No 8, pp 411 – 416 [5] https://www.investopedia.com/articles/economics/08/past-recessions.asp [6] Donny Susilo (2018) The Impact of Foreign Direct Investment on Economic Growth (a Causal Study in the United States) [7] Shrikant Panigrahi, Noor Azizan, Shahryar Sorooshian, Prabha Thoudam (2020) Effects of Inflation, Interest, and Unemployment Rates on Economic Growth: Evidence from ASEAN Countries FE64 - Group 11 12 Econometrics Asignment APPENDIX Model [6] Representations: Estimation Command: LS LOG(GPDI) C LOG(FDI) LOG(TAX) U RECESS U*RECESS Estimation Equation: LOG(GPDI) = C(1) + C(2)*LOG(FDI) + C(3)*LOG(TAX) + C(4)*U + C(5)*RECESS + C(6)*U*RECESS Substituted Coefficients: LOG(GPDI) = 1.14978960124 + 0.149019780837*LOG(FDI) + 0.642427754589*LOG(TAX) - 0.029781326913*U - 0.149763668376*RECESS + 0.0185807532201*U*RECESS Estimation outputs: Dependent Variable: LOG(GPDI) Method: Least Squares Date: 11/04/23 Time: 12:04 Sample: 2001Q1 2023Q2 Included observations: 90 Variable Coefficient C 1.149790 LOG(FDI) 0.149020 LOG(TAX) 0.642428 U -0.029781 RECESS -0.149764 U*RECESS 0.018581 R-squared 0.973582 Adjusted R-squared 0.972010 S.E of regression 0.045971 Sum squared resid 0.177516 Log likelihood 152.5781 F-statistic 619.1323 Prob(F-statistic) FE64 - Group 11 Std Error t-Statistic 0.211105 5.446542 0.041818 3.563555 0.065131 9.863644 0.003802 -7.833050 0.036041 -4.155324 0.005561 3.341525 Mean dependent var S.D dependent var Akaike info criterion Schwarz criterion Hannan-Quinn criter Durbin-Watson stat Prob 0.0000 0.0006 0.0000 0.0000 0.0001 0.0012 7.950105 0.274774 -3.257291 -3.090637 -3.190087 0.512135 0.000000 13