Tugas Komputasi Statistika

Klasterisasi Kecamatan Berdasarkan Karakteristik Wilayah di Lamongan Menggunakan K-Means Clustering

Dataset yang digunakan adalah Wisconsin Breast Cancer Dataset yang berisi 569 observasi dengan 30 fitur numerik hasil pengukuran sel tumor payudara. Variabel dependen yang digunakan adalah radius_mean (rata-rata radius tumor), sedangkan variabel independen meliputi texture_mean, smoothness_mean, compactness_mean, dan concavity_mean.

To answer the R&D expenditure and firm size analysis task concisely, I first simulate the provided data in R by running the code to create the df_firms dataframe. For visualization, I use ggplot2 to generate a scatter plot of Total_Assets vs. RD_Expenditure, adding a linear trend line to highlight the relationship. In diagnosis, I fit a simple OLS model with lm(RD_Expenditure ~ Total_Assets), then check residuals for normality using a Q-Q plot and Shapiro-Wilk test, and for homoscedasticity with a residuals-vs-fitted plot and Breusch-Pagan test. For transformation, I apply boxcox() from the MASS package on the model to identify the optimal lambda (likely near 0, suggesting a log transform). Finally, I refine by transforming RD_Expenditure based on lambda, re-fit the OLS model, compare summaries and diagnostics (e.g., improved R-squared and test p-values), structure everything in an R Markdown file.