diff --git a/.RData b/.RData
index 1413c4a..d9d26e0 100644
Binary files a/.RData and b/.RData differ
diff --git a/020325_CONTRIBUTING_commit_topic_model.rda b/020325_CONTRIBUTING_commit_topic_model.rda
new file mode 100644
index 0000000..dd3e05f
Binary files /dev/null and b/020325_CONTRIBUTING_commit_topic_model.rda differ
diff --git a/topic-outcome-models/readme_topic_outcome_model.R b/topic-outcome-models/readme_topic_outcome_model.R
index 1f888e3..6b4100c 100644
--- a/topic-outcome-models/readme_topic_outcome_model.R
+++ b/topic-outcome-models/readme_topic_outcome_model.R
@@ -6,7 +6,7 @@ library(stringr)
 readme_count_data_filepath <- "/mmfs1/gscratch/comdata/users/mjilg/govdoc-cr-data/final_data/README_weekly_count_data.csv"
 readme_count_df = read.csv(readme_count_data_filepath, header = TRUE) 
 
-readme_topic_dist_filepath <- "text_analysis/020125_README_file_topic_distributions.csv"
+readme_topic_dist_filepath <- "text_analysis/020325_README_file_topic_distributions.csv"
 readme_topics_df = read.csv(readme_topic_dist_filepath, header = TRUE) 
 
 window_num <- 5
@@ -22,18 +22,18 @@ summed_data <- readme_count_df |>
   summarise_at(vars(commit_count), list(summed_count=sum))
 
 readme_topics_df <- readme_topics_df |>
-  mutate(project_id = sapply(str_split(filename, "_hullabaloo_"), `[`, 1)) |> 
+  mutate(project_id = sapply(str_split(filename, "_hullabaloo_"), `[`, 1)) |>
   mutate(project_id = ifelse(filename=="jaraco_keyrings.alt_hullabaloo_README.rst", "jaraco_keyrings.alt", project_id)) |>
   mutate(project_id = ifelse(filename=="_vcr_vcr_README.md", "vcr_vcr", project_id)) 
   
-filtered_topics <- readme_topics_df |>
-  filter(project_id %in% summed_data$project_id)
-
+#loss of jaraco_keyring, though jaraco keyrings.alt is represented
 merged_df <- inner_join(summed_data, readme_topics_df, by="project_id")
 
-multiple_mappings <- merged_df %>%
-  group_by(project_id) %>%
-  filter(n() > 1) %>%
-  ungroup()
-
 merged_df$logged_commits <- log1p(merged_df$summed_count)
+
+library(MASS)
+commit_outcome_model <- glm.nb(logged_commits ~ 0 + t0 + t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10, data=merged_df)
+qqnorm(residuals(commit_outcome_model))
+summary(commit_outcome_model)
+
+saveRDS(commit_outcome_model, "020325_CONTRIBUTING_commit_topic_model.rda")