The Impact of Machine Learning on Economics by Susan Athey

In her article "The Impact of Machine Learning on Economics" (2018), Susan Athey provides an in-depth exploration of how machine learning is poised to revolutionize the field of economics and public policy. She emphasizes that this technology opens up new opportunities to address complex economic problems, particularly in prediction, causal analysis, and personalized policy design.

Athey explains that machine learning, with its ability to process large and complex datasets, excels at generating highly accurate predictions. This capability is invaluable in forecasting various outcomes, such as consumer behavior, household spending patterns, or labor market trends. However, she stresses that the power of machine learning extends beyond prediction—it also aids causal analysis, which involves understanding cause-and-effect relationships. For instance, machine learning can be used to evaluate the impact of subsidies on energy consumption or to determine how job training programs influence income growth.

One of Athey's key contributions is clarifying the distinction between prediction and causal analysis. Prediction focuses on forecasting future outcomes based on historical data patterns, while causal analysis aims to understand the specific effects of a policy or intervention. Machine learning, when paired with traditional statistical methods, can address challenges in both areas, offering innovative solutions to longstanding economic questions.

Athey highlights several machine learning methods relevant to economics, such as random forests and causal forests. These approaches not only enhance predictive accuracy but also allow for deeper insights into how policies might affect different segments of the population. For example, causal forests can help policymakers design interventions tailored to vulnerable groups, such as low-income households or at-risk students.

She provides practical examples of how machine learning can be applied in public policy. Governments can use this technology to target social assistance to the most disadvantaged populations, improve educational outcomes by identifying students at risk of dropping out, or design data-driven health policies to boost vaccination compliance. With its ability to personalize policies, machine learning has the potential to make public programs more efficient and effective.

Despite its promise, Athey acknowledges the challenges associated with applying machine learning in economics. One major issue is data bias—if the data used to train algorithms are biased, the resulting predictions and decisions will reflect these biases. Additionally, machine learning models are often seen as "black boxes," making it difficult to interpret their results, which can hinder their application in public decision-making. Ethical and privacy concerns are also significant, particularly when using big data to inform policies.

Through this article, Susan Athey underscores the importance of collaboration between economists, data scientists, and policymakers to fully harness the potential of machine learning. By integrating this modern technology with traditional economic approaches, machine learning can become a powerful tool for advancing economic analysis and implementing more effective public policies. The article represents a major step forward in bridging the gap between economics and technological innovation, demonstrating how these advancements can be leveraged to create meaningful and positive societal impacts. 

Review

In her article "The Impact of Machine Learning on Economics" (2018), Susan Athey presents a pivotal discussion on how machine learning is transforming economics and public policy. She highlights that this technology goes beyond improving prediction accuracy; it also supports causal analysis—uncovering cause-and-effect relationships, such as the impact of subsidies on consumption or job training on income. Using tools like causal forests, Athey emphasizes the potential for personalized policies to address the specific needs of vulnerable groups, such as at-risk students or low-income households.

Athey also acknowledges key challenges, including data bias, privacy concerns, and the "black box" nature of machine learning algorithms, which can limit their interpretability. While the article provides examples of applications in education, healthcare, and poverty alleviation, it lacks empirical studies to solidify its claims. Furthermore, its technical depth might be less accessible to non-technical readers.

Despite these limitations, the article remains a significant contribution, bridging the fields of economics and modern technology. By incorporating concrete evidence and a stronger focus on ethics, it could serve as a strategic guide for interdisciplinary collaboration needed to create impactful, data-driven policies.

Read more...

Understanding the Dynamics of Indonesia's Economy: Challenges and Solutions Amid Global Uncertainty

By Sanjoyo

In recent months, Indonesia's economy has shown various intriguing dynamics. Despite global challenges, Indonesia has managed to maintain stable economic growth while leveraging available opportunities. Let's take a closer look at the latest developments, notable issues, and potential solutions to address these challenges.

Stable Economic Growth. Indonesia's economy has demonstrated stable growth, with GDP projected at around 5% in 2023. This growth is supported by strong domestic consumption and continuous investment, particularly in infrastructure and technology sectors. Government policies, such as the Omnibus Law, have also provided a boost to the investment climate, although some challenges remain to be addressed.

Current Notable Issues

1. Food and Energy Prices. One of the main concerns is the rising prices of food and energy. The surge in rice and cooking oil prices is caused by global supply chain disruptions and climate changes affecting domestic production. This directly impacts people's purchasing power and economic stability.

2. Global Uncertainty. Global uncertainties, particularly due to the Russia-Ukraine war and international trade tensions, have negatively affected trade and investment. Fluctuations in commodity prices, such as oil and gas, also significantly impact Indonesia's economy.

3. Health and Post-Pandemic Economic Recovery.  Although COVID-19 cases have declined, economic recovery still faces various challenges. The tourism and MSME sectors require full support to recover and contribute optimally to the economy.

4. Digitalization and Economic Transformation. Increasing adoption of technology and digitalization is a primary focus for both the government and the private sector. This transformation is crucial for improving efficiency and competitiveness in the global economy.

Propose Policies Direction. 

To address these challenges, several strategic steps can be taken:

1. Price and Subsidy Policies. The government needs to strengthen price and subsidy policies to maintain food and energy price stability. Support for farmers and the agricultural sector should also be increased to boost domestic production.

2. Economic Diversification. Reducing dependency on commodities by strengthening the manufacturing and service sectors can enhance economic resilience. Investment in technology and innovation should be encouraged to create a more dynamic and competitive economy.

3. Infrastructure Strengthening. Continuous improvement of physical and digital infrastructure is necessary. Major infrastructure projects, such as the construction of toll roads, ports, and internet networks, need to be expedited to support long-term economic growth.

4. Support for MSMEs and Tourism. Expanded assistance and incentives for MSMEs and the tourism sector are essential. Training and financial access for MSMEs should be increased to support economic recovery and growth.

5. Global Risk Management. Developing risk mitigation strategies against global uncertainties is crucial. Diversifying export markets and strengthening international cooperation can effectively reduce the negative impacts of global uncertainties.

With appropriate policies and support from various stakeholders, Indonesia can continue to grow and overcome existing economic challenges. It is important to continuously monitor global and domestic developments and adjust policies as needed. By doing so, Indonesia can maintain its economic stability and progress amid global uncertainty.

Read more...

Indonesia's Future Economy

Indonesia's future economy is poised for significant growth and transformation, with the resource sector playing a crucial role. As one of the world's largest archipelagic nations, Indonesia boasts abundant natural resources, including oil, natural gas, coal, minerals, and a vast expanse of forests. Historically, these sectors have been major contributors to the country's economy, generating significant revenue and employment opportunities. 

However, Indonesia is increasingly recognizing the need to diversify its economy and reduce its dependence on resource extraction. While the resource sector will likely continue to play a vital role in the economy for the foreseeable future, there is a growing emphasis on sustainable development and leveraging these resources more efficiently and responsibly. 

Here are some key aspects of Indonesia's future economy and the role of resource sectors:

Diversification: Indonesia aims to diversify its economy by promoting other sectors such as manufacturing, tourism, technology, and services. This diversification strategy is intended to reduce the economy's vulnerability to fluctuations in commodity prices and global demand for resources.

Sustainable Development: There's an increasing focus on sustainable development practices within the resource sectors. This includes efforts to minimize environmental degradation, promote renewable energy sources, and adopt responsible mining practices. Initiatives like the Sustainable Development Goals (SDGs) provide a framework for guiding these efforts.

Infrastructure Development: The development of infrastructure is essential for unlocking the full potential of Indonesia's resource sectors. Investments in transportation, energy infrastructure, and telecommunications are critical for improving accessibility to remote resource-rich areas and facilitating efficient resource extraction and distribution.

Technology and Innovation: Embracing technological advancements and innovation is crucial for enhancing productivity and competitiveness in the resource sectors. Technologies like automation, artificial intelligence, and remote sensing can improve efficiency, safety, and environmental sustainability in mining, oil, and gas exploration, and forestry operations.

Value-Added Processing: Instead of solely exporting raw materials, there's a push to promote value-added processing within Indonesia. This involves developing downstream industries to refine and process raw materials domestically, thereby capturing more value and creating higher-skilled jobs.

Government Policies and Regulations: The Indonesian government plays a central role in shaping the future of the resource sectors through policies, regulations, and incentives. It needs to strike a balance between attracting investment, ensuring environmental protection, and maximizing benefits for local communities.

International Partnerships: Collaboration with international partners and investors can bring in expertise, technology, and capital to support the development of Indonesia's resource sectors. However, it's essential to negotiate fair deals that prioritize the long-term interests of Indonesia and its people.

In summary, while the resource sectors will remain significant drivers of Indonesia's economy, the country is actively pursuing strategies to diversify its economy, promote sustainability, and maximize the value derived from its abundant natural resources. Balancing economic development with environmental and social concerns will be crucial for ensuring a prosperous and equitable future for Indonesia. 

Read more...

Indonesia Economic Growth

• In the 2015-2019 period, 11.88 million jobs were created. In August 2019, TPT fell to 5.28 percent compared to last year's 5.34 percent.

• The tourism sector, as one of the drivers of the Indonesian economy, is experiencing good progress. According to The Travel & Tourism Competitiveness Report released by the WEF (World Economic Forum), the ranking of Indonesia's tourism competitiveness index in the world rose to 40 in 2019 from 42 in 2017.

• The interest of foreign tourists or tourists to Indonesia is increasing, it is shown that November 2019 data has increased by 11.55 percent compared to the number of visits in November 2018.

Read more...

Monetary Policy and Economic School of Thought

Monetary Policy and Economic School of Thought. The influence of monetary policy on output and prices is a long debate concerning both theoretical and empirical terms. This is not detached from the economic development school of thought from start Classical, Neo-classical, Neo-classical synthesis, New Classical and New Keynesian.

In Classical view, money affect only the price and not to the output. By using analysis general equilibrium, included money in the model showed the money neutrality that money does not affect on market equilibrium. On the other hand, the Keynesian view of money affects prices and output because of the price rigidity and involuntary unemployment. This view is modeled on the IS-LM for the equilibrium of money market and goods market and disequilibrium on the labor market.

In the 1960s there consensus view that the money may affect output and prices in the short term (Neoclassical Synthesis). In that period, structure of the labor market is replaced with Phillips curve as aggregate supply. Neoclassical Synthesis model explained that the occurrence of rigidity prices and wages because of the assumptions in determining the behavior of company that is the price mark-up of wages. Although the real wage is flexible, but the pricing behavior conducted in the mark-up so lead to occurring rigidity wages and prices and then money supply affect real output and prices.

Expectation economic agent face economic uncertainty will influence macroeconomic. Two important hypothetical of expectation in the economy are rational expectation and adaptive expectation. Milton Freidman (1957) introduced the adaptive expectation that the expectation economic agents formed by observations of inflation at this time. Phenomenon of the Phillip curve was challenged by Friedman points out that the argument only unanticipated inflation are affecting unemployment. He emphasize on the importance of expectation on the aggregate supply so that revised Phillips curve as expectation-augmented Phillips curve.

On 70’s, it is difficult period for the Keynesian. Lucas (1976) and Sargent-Wallace (1975) introduced the rational expectation that assume economic agents use all relevant information to establish expectation or forecast economic variables in the future. So that monetary policy and fiscal policy affects inflation, than expectation inflation also depend on effect those policies. Thus, changes in monetary and fiscal policies affect the changes expectation agent economy. So, the policy evaluation must consider the effects of expectation economic agents.

Lucas (1976) criticize the results of parameter estimation econometric model that is not stable because occurring the changes policy maker behavior, and than expectation of the private agent will also be changed then it will affect the parameters in the econometric model. This critique affect on two, the revised macroeconomic model with rational expectation of entering and strengthening macroeconomic model with micro foundation.

At the 80’s Classical school of thought was extremely dominant. In the New Classical paradigms, Kydland - Prescott (1982) introduced the real business cycle theory (RBC), which begins with microeconomic assumption of household consumption preference, the production firm and market structures. With the intertemporal optimization of consumption of households and future profit of firms and the market is competitive then the solution obtained by dynamic general equilibrium model. They succeeded in making data replication USA. RBC model assume output is always in the natural level of output and all of output fluctuations are the movement of natural level of output itself. The cause of output fluctuations in Prescott point of view is a shock or a change in technology. Similarly, in the RBC model change in money supply does not affect output.

After the 80’s, research on RBC develop in many models. Debate on technology shock provides inspiration for researchers to develop various models incorporate various aspects, among others; oil shock, fiscal shock, monetary model, and the multiple equilibrium model (Rebelo, 2005).

The latest research on the RBC model related to monetary policy is to include elements of nominal wage and price rigidity in the model, so that changes in money supply can affect output. This model, known as Dynamic Stochastic General Equilibrium (DSGE) model. Some researchers Christiano, Eichenbaum and Evans (2003), Woodford (2003), Smets and Wouters (2004); and Laxton and Pesenti (2003) build and estimate DSGE model based RBC with assumptions nominal rigidities in wage and price, including assumption imperfect competition in market labor market and product market.

Another mainstream New Keynesian is the improvement of the Neo-Classical synthesis with incorporate aspects of the rational expectation and strengthening the micro foundations. However, the Keynesian economists still believe the existence of imperfect markets and nominal rigidity can lead to fluctuations (deviation) of output from natural output. Fischer (1977) and Taylor (1980) argued that the occurrence of the nominal rigidity caused staggering of wage and price decisions by firms. The existence of staggering in wage and price lead to adjustments on price level slowly so that changes in aggregate demand impact on output fluctuations.

In New Keynesian paradigms, the economists [Gali and Gertler (1999) and Gali et al. (2001), Roberts (2001), Fuhrer (1997); Linde (2005)] has to learn how to develop a simple model, related, and structural that could explain mechanisms transmission of monetary, especially through the interest rate and the impact on inflation and output. This model is known as model New Keynesian Small Macroeconomics (NKSM) with approach dynamic stochastic general equilibrium that contain aspects expectation and also solid with micro foundation. This simple model is also containing the aggregate demand, price-setting (Phillips) curve, and the reaction of an interest rate policy to output and inflation. This model to realize the basic principle of the role of monetary policy instruments through the nominal interest rate to inflation stabilization.

Technically DSGE models have weaknesses in terms of technique calibration that difficult to create the replication of data in accordance with the actual data, but the advantage that the DSGE model parameter is the "deep parameters" (parameters for the micro variables). While NKSM have benefits to explain economic conditions simpler, but the weakness is difficult to get the relationship between variables significantly because of the unobserved variables or serial correlation.

Paper in Indonesia Language.
Keyword: Classical, DSGE, Economic, IS-LM, Neo-classical, Neo-classical synthesis, New Classical and New Keynesian, Phillips curve, Price rigidity, Rational Expectation, RBC, wage rigidity.

Read more...

The Economist on Indonesia

The Economist are quite optimistic on Indonesia:
The data suggest the fourth-quarter slowdown in Indonesia was much less pronounced than elsewhere in South-East Asia. Economic growth for 2008 as a whole is likely to exceed 6%. The 2008 budget deficit was 0.1% of GDP and the government has earmarked $3.5 billion to spend on tax breaks and infrastructure projects. In late 2008 the currency, the rupiah, lost a fifth of its value against the dollar, but the slide has halted. The cost of insuring Indonesian government bonds against default has come down sharply. Inflation, still running at an annual rate of 11%, is falling. The central bank cut interest rate by one-half of a percentage point, to 8.75%. Most banks are healthy. Moody’s, a credit-rating agency, gave Indonesia a “stable” outlook in its annual report this week, expecting the authorities to manage the impact of the crisis competently.

Click here to read complete article.

Read more...

Currency Crisis Effect on the Stock Market: A Case Study in Indonesia

Currency Crisis Effect on the Stock Market: A Case Study in Indonesia. This essay has analyzed the Indonesian currency crisis by empirically examining relationships between the composite share price index and sectoral stock indices of the Jakarta Stock Exchange and the exchange rate. The results show that changes in composite price index and property and real estate, financial indices provided early indication of the currency crisis when the central bank applied a managed float regime. The VAR test shows that these stock indices (in differences) caused exchange rate changes before the crisis period.

There has been a strong causality relationship from the rupiah exchange rate to the composite share price index and all the stock indices during the post crisis period. All stock market indices can be explained by the exchange rate changes. The tradable goods producers such as mining and manufacturing have positive sensitive to exchange rate changes. They become better off when the exchange rate depreciates. On the other hand, non-tradable goods producers such as property and real estate and infrastructure will become worse off when the exchange rate depreciates.

The causality relationship between the exchange rate and the stock market indices disappeared during the peak crisis period. Many factors influenced the exchange rate, such as a social and political instability and a loss of confidence by investors.

This study has implications for monitoring financial markets. Currently, the monitoring practice for the financial sector is based on a portfolio approach, often relying on low frequency data, due to the reporting practices of financial entities. However, high frequency financial indices such as stock indices can supplement some deficiencies of the conventional method, as the stock indices such as the composite share price index, financial, property and real estate indices showing early indicator currency crisis, especially in the pre-crisis period.
Finally, for further research we can relax the assumption of a constant interest rate and develop the model considering the time-varying interest rate. This is because we should consider the variation of domestic interest rates in order to improve the goodness of fit of the model.

Complete paper in English

Summary paper in Indonesia language.

Keyword: Composite Index, Currency Crisis, Exchange rate, Financial, Jakarta Stock Index, Rupiah, Stock Market

Read more...

PANEL UNIT ROOT TEST

In the last decade, the issue of unit root test for heterogenous panels has attracted many academic researchers. In principle the application of the panel data unit root test is intend to increase the power of test by increasing number of sample. Increasing amount of sample can be done by increasing the number of cross sectional data and the number of time series data. The problem emerge in panel data are the issue of structural change when using long data series or contain heterogeneity when using cross sectional data. The famous example unit root test for homogenous panel was Summer and Heston (1991) using a panel data set covering a variety of industry, region, various country with a long period of time.
Unit root test has been developed by Quah (1992.1994), Levin and Lin (1993) for homogenous panels. That testing the unit root can not accommodate heterogeneity between groups, such as the unique influence of individuals (individual special effects) and a different pattern of residual serial correlations. Test statistics that proposed by Quah, Levin and Lin can more be used with the conditions for the existence of specific individual effects and also heterogeneity across groups and then requires N / T -> 0 and two N (cross section dimension) and T (time series dimension) toward unlimited.

Pesaran and Smith (1995), and Pesaran, Smith and Im (1996) showed that the inconsistencies in the estimation model dynamic heterogeneous panels. Furthermore, based on the paper, the Im, Pesaran and Shin (2002) introduced the unit root test with dynamic heterogeneous panels. In general, the unit root test with dynamic heterogeneous more used compared with the homogenous dynamic. Im, Pesaran and Shin (IPS) framework using the likelihood procedure based on an alternative test average unit root test statistics in each individual group for the panel. IPS was testing based on the average (augmented) Dickey Fuller (1979), which refers to the t - test bar. Such as procedures performed by Levin and Lin, unit root test done by the IPS is to consider the characteristics of serial correlation dynamics and heterogeneity residues for each panel group. Statistics (IPS) is indicated in the convergence of the standard normal probabilities in line with sequential T to unlimited number, and followed by the N to unlimited number, where T is the time series dimension and N is the cross sectional dimension. The diagonal convergence between T and N to unlimited number, while NT -> k, where k is a constant non-negative limited number. In special cases, where the residual of the individual DF Regression are serially correlated, then Z ~ tbar which is a modified t-stat will distributed with the normal standard at the time of N → ∞ and T fixed, so that the length of T> 5 for the regression with the intercept and DF T> 6 for DF regression with intercept and linear time trends. Next, the test was also developed to test how T and N fixed with the average DF. Simulation results that with the big ordo of the ADF regression; the performance of the limited sample t-bar test is very satisfactory and gives better results than Levin-Lin (LL) test. Therefore, in this paper will attempt to simulate formulas and procedures of Pesaran.

Complete Paper in Indonesia Language.
Summary Paper in Indonesia Language.

Read more...

Estimation Non Linier Model with Genetic Algoritma

In general, estimates in Model Non Linier method using OLS (Ordinary Least Square) or ML (Maximum Likelihood) with conventional algorithms method such as Gause-Newton; Rhapson-Newton, Levenberg-Marquardt; Berndt, Hall, Hall & Hausman or the quadratic Hill-Climbing. These Algorithms will not produce a global minimum / maximum. In this paper will explain the new approach, namely Genetic Algorithm to ensure global maximum/ minimum. Monte Carlo simulation is used to guarantee the results Robusness estimates. Computing used MATLAB.

Download if you want to get the full paper: Genetic Algoritma.pdf

Read more...

ROLE THE PUBLIC SECTOR IN ACCUMULATION OF HUMAN CAPITAL AND CAPACITY RESEARCH & DEVELOPMENT

In the theory of economic growth, sources of economic growth, sources of growth - comes from the ability of a country in developing potential resources. Quality and the greater the higher the quality of resources, it also has a greater potential to increase a country's economic growth. Factors that are important in sources of growth are natural resources, capital, saving, and development of technology. Property natural resources would help the economy of a country, although not enough if not supported by the skill of exploration for natural resources.

Both capital and saving is also a factor of production as the dominant element of economic growth for the future. Similarly, the development of technology can be widely accepted as a source of economic growth. This is because the technology that allows for manufacturers to produce more with the same input level. The development of technology depends on the ability of science and the quality of education of a country and how much attention on research and development.

Results of empirical studies of economic growth showed that the relationship is strong economic development of a country with a capacity of human capital the country. However, the dynamic relationship between economic growth with human capital and research & development can be explained since the 1980 when Romer and Lucas describe the relationship with the growth model endogenous or new growth theory.

In this paper will try to explain briefly the history of development of the economics of growth theory and briefly review the core of neoclassical model and endogenous growth models. Then, this paper will explain the role of accumulation of human capital and research & development in the economic development. Finally, this paper will review the role of the public sector and policy implications in the process of accumulation of human capital and investment of R & D to contribute to economic growth.

Summary paper in Indonesia language.

Complete paper in Indonesia language

Read more...

NonLinier Estimation using OLS and Max Likelihood

Paper will report the results experiment model nonlinear to estimate production function Cobb-Douglas and CES using the Least Square method Nonlinear and Non-Linier Maximum Likelihood. Model estimation method linier used non-conventional approach Algorithm Gause-Newton; Rhapson-Newton, Levenberg-Marquardt; Berndt, Hall, Hall & Hausman or the quadratic Hill-Climbing. In this paper will describe the approach. Monte Carlo simulation is used to guarantee the results Robusness estimates. Computing used MATLAB.

Download if you want to complete any posts [Non Linier.pdf] dan [Lampiran]

Read more...

Ordinary Least Square Estimation in Linier Model and Monte Carlo Simulation

Ordinary Least Square Estimation in Linier Model and Monte Carlo Simulation. In general methods of estimation in Model Linier used OLS (Ordinary Least Square) or ML (Maximum Likelihood). In this Paper describes theoretically how the estimate methods are. Monte Carlo simulation is used to guarantee the results robusness estimates. Computing used MATLAB. Econometric. Ekonometrik

Download if you want to get the full paper: (1)cover.pdf; (2)daftar-isi.pdf; (3)isi.pdf

Keyword: Econometric, Ekonometrik, Estimation, MATLAB, Maximum Likelihood, Monte Carlo, OLS, Ordinary Least Square, Simulasi, Simulation.

Read more...

New Keynesian Macroeconomics Model for Indonesia

In the recent years many academic interested in New Keynesian Small Macroeconomic model that is developed in the mid of 90's as a response from Lucas Critique on macroeconomic Keynesian (IS-LM) in early 80's. Repairs carried out on Keynesian model with a base to build models based on micro foundation (house hold and firm optimization) and incorporate aspects of rational expectation.

In the beginning New Keynesian Small Macroeconomic model was built for a closed economy model for the small country (in the sense that the influence of a small country of the world economy). Then around the beginning of the year 2000’s model developed for the open economy. Main core (standard model) this New Keynesian Small Macroeconomic model is the three equation such as; the aggregate demand equation (which was formed from the optimization intertemporal consumption of house hold); the aggregate supply (which was formed Maximization of discounted future profit company); and the rule of Monetary (Taylor rule).

At this time the researchers continue to develop based on the standard model with a special specifications are: the aggregate supply; add capital factors; add a factor in oil prices. Similarly, many models have been used by the Central Bank in some countries and used by the IMF as forecasting and policy analysis system Model (FPAS) to evaluate the macroeconomic state of the member-countries.

Download if you want to complete any posts [New Keynesian.pdf]

Read more...