Best Claude prompts for Budget Analysts

Given a dataset of historical budget allocations and actual expenditures, develop a regression model to forecast future budget needs. The dataset includes variables such as department, funding source, and economic indicators. Use Python and scikit-learn to implement the model, and evaluate its performance using metrics such as mean absolute error and coefficient of determination. Provide a detailed report on the model's strengths, weaknesses, and recommendations for improvement. Assume a sample size of 500 observations and a 80/20 train-test split.

Conduct an audit of an existing ETL pipeline that extracts data from a Snowflake database, transforms it using Python, and loads it into a Tableau dashboard. Evaluate the pipeline's performance, data quality, and compliance with organizational standards. Identify potential bottlenecks, data inconsistencies, and areas for optimization. Provide a detailed report on the audit findings, including recommendations for improving data accuracy, reducing processing time, and enhancing overall pipeline efficiency. Assume the pipeline processes 100,000 records daily.

Analyze a dataset of budget variances across different departments and funding sources, and provide a statistical summary of the findings. Calculate metrics such as mean, median, standard deviation, and interquartile range for the variance amounts. Use SQL to query the data and Python to perform the statistical analysis. Visualize the results using a histogram and box plot to identify patterns and outliers. Provide a detailed report on the insights gained from the analysis, including recommendations for addressing significant budget variances. Assume a dataset of 10,000 observations.

Optimize a set of SQL queries used for budget reporting, with the goal of reducing query execution time and improving data retrieval efficiency. The queries involve complex joins, subqueries, and aggregations across multiple tables in a Snowflake database. Use the EXPLAIN and ANALYZE commands to evaluate query performance, and apply optimization techniques such as indexing, caching, and rewriting queries to improve performance. Provide a detailed report on the optimization results, including before-and-after comparisons of query execution times and recommendations for further improvement. Assume a dataset of 1 million records.