In this draft analysis, I explore the relationship between CRE and rent prices in DC.

We present a novel approach to agent-based modeling by replacing traditional utility-maximizing agents with Large Language Model (LLM) agents that make human-like residential decisions. Using the classic Schelling segregation model as our testbed, we compare three agent types: (1) traditional mechanical agents using best-response dynamics, (2) LLM agents making decisions based on current neighborhood context, and (3) LLM agents with persistent memory of past interactions and relationships. Our results reveal that LLM agents achieve complete convergence (100%) while mechanical agents only converge 50% of the time. Standard LLM agents converge in 99±9 steps compared to 187 steps for mechanical agents when they do converge. Memory-enhanced LLM agents demonstrate the fastest convergence at 84±14 steps—a 2.2× improvement. Both LLM variants achieve similar final segregation levels to mechanical agents (~55% vs 58% like-neighbors) but with significantly reduced extreme segregation, with memory LLM agents showing a 53.8% reduction in “ghetto” formation (p=0.018). These findings suggest that incorporating human-like decision-making through LLMs can produce more stable and realistic dynamics in agent-based models of social phenomena, with important implications for urban planning and policy analysis.

This presentation showcases a predictive text application developed using R, Shiny, and NLP techniques. The app uses a backoff n-gram language model to predict the next word in a user’s input phrase based on statistical patterns learned from a large corpus of blogs, news articles, and tweets

We present a novel approach to agent-based modeling by replacing traditional utility-maximizing agents with Large Language Model (LLM) agents that make human-like residential decisions. Using the classic Schelling segregation model as our testbed, we compare three agent types: (1) traditional mechanical agents using best-response dynamics, (2) LLM agents making decisions based on current neighborhood context, and (3) LLM agents with persistent memory of past interactions and relationships. Our results reveal that LLM agents achieve complete convergence (100%) while mechanical agents only converge 50% of the time. Standard LLM agents converge in 99±9 steps compared to 187 steps for mechanical agents when they do converge. Memory-enhanced LLM agents demonstrate the fastest convergence at 84±14 steps—a 2.2× improvement. Both LLM variants achieve similar final segregation levels to mechanical agents (~55% vs 58% like-neighbors) but with significantly reduced extreme segregation, with memory LLM agents showing a 53.8% reduction in “ghetto” formation (p=0.018). These findings suggest that incorporating human-like decision-making through LLMs can produce more stable and realistic dynamics in agent-based models of social phenomena, with important implications for urban planning and policy analysis.

Trabajo Practico de Series de Tiempo para la materia de Econometría

Penelitian ini menggunakan data sekunder dari BPS tahun 2024, yang mencakup AHH, RLS, dan HLS untuk setiap provinsi di Indonesia. Dengan menerapkan clustering menggunakan metode K-Means sangat relevan untuk mengelompokkan provinsi berdasarkan kemiripan nilai ketiga indikator tersebut. Dengan pendekatan ini, pola pembangunan manusia antar wilayah dapat diidentifikasi secara lebih jelas, sehingga kebijakan pembangunan dapat difokuskan untuk mengurangi kesenjangan antarprovinsi dan meningkatkan kualitas hidup masyarakat secara merata.