Kyumin Kim | Technical Showcases

I specialize in quantifying risk and uncertainty through dynamic optimization, econometrics, and machine learning—with applications in natural resource and environmental systems and in business decision-making under uncertainty.

This page is my technical portfolio: self-contained examples of the methods I use across research and applied work. Each one is a focused, hands-on demonstration, intentionally built as a compact proof of concept rather than a full research paper, scoped to isolate a single method or idea and show it working. Together they cover dynamic programming, structural econometrics, machine learning, and Monte Carlo–style simulation, and they show how I translate quantitative methods into working code that solves a concrete problem.

Business Decision Making under Uncertainty

An integrated decision-making process under uncertainty: combining dynamic optimization with prediction and forecasting from machine learning and econometrics.

The Value of Information from a Forecast: Inventory, Demand Uncertainty, and an Imperfect ML Signal: Series 1Dynamic programming, value of information, Bayesian updating, inventory managementML demand forecasts, production under uncertainty, continuous-state VFI, business decision making

Dynamic Optimization, Uncertainty and Risk Management in Natural Resource Economics

Exploring dynamic programming approaches and risk management strategies in fisheries decision-making under uncertainty.

Solving a Dynamic Problem: Fisheries Management with Dynamic Programming: Series 1Dynamic programming, Bellman equation, value function iterationDynamic optimization, Fisheries management, Sustainable resource use

Optimal Fisheries Management Under Uncertainty: Using Stochastic Dynamic Programming: Series 2Stochastic dynamic programming, Monte Carlo integration, climate riskUncertainty handling, Risk management, Stochastic optimization

Optimal Tree Harvesting under Price Uncertainty: A Small Dynamic-Programming Example by Markov-Transition Matrix (MTM): Series 3Markov transition matrix, price uncertainty, optimal harvestingPersistent uncertainty, Threshold policies, Natural resource economics

Econometrics, Machine Learning, and Prediction

Econometrics

Structural Estimation via NFXP: A Step-by-Step Guide to the Rust (1987) Bus Engine Model: Series 1Dynamic discrete choice, Nested Fixed-Point algorithm, structural parametersRust (1987), Bellman equation, Gumbel errors, MLE, dynamic discrete choice model, structural econometrics

Machine Learning and Economics

A Friendly Introduction to Machine Learning: Purpose, Validation, and the Bias–Variance Tradeoff: Series 1Machine learning fundamentals, cross-validation, bias-variance decompositionPrediction vs. causal inference, K-fold CV, Model evaluation

A Friendly Introduction to Shrinkage Estimators: Ridge, LASSO, Elastic Net, and the Adaptive LASSO: Series 2Shrinkage methods, regularization, variable selection, oracle propertyRidge regression, LASSO, Elastic Net, Adaptive LASSO

Machine Learning Series 3: Decision Trees and Random Forests: A Friendly Introduction: Series 3Decision trees, random forests, ensemble methods, variable importanceCART, Bootstrap sampling, Bagging, Medical expenditure prediction

Machine Learning Series 4: Random Forest Hyperparameters and Practical Grid Search: Series 4Hyperparameter tuning, grid search, bias-variance tradeoff, model selectionntree, mtry, maxnodes, OOB error, Test RMSE, MEPS 2003

Sustainability, Emission, and Uncertainty

A series exploring the intersection of fisheries management, carbon emissions, and economic uncertainty through bioeconomic modeling and Monte Carlo simulation.

Optimal Fishing Management Under Emission Cost in Korean Fisheries: Series 1Optimal fishing efforts, emission cost, and bioeconomic analysisEconomic optimal decision, Fisheries management, GHG Emission cost

Fisheries and Emission Cost: Addressing Economic Uncertainty through Monte Carlo Simulation: Series 2Economic optimal decision, Economic uncertainty, Monte Carlo simulationEconomic optimal decision, GHG emission cost under uncertainty, Monte Carlo simulation