Julia provides incredible flexibility for migrating tedious data processing and decision/risk modeling that happens currently in Excel but no packages existed to automate or simplify the processing of simulation results. Inspired by the many Excel add-ins that were created to automate the many steps in the Monte-Carlo Simulation algorithm (including correlation and results visualization), we created a similar time-saving toolkit for Julia called MCHammer.
MCHammer provides all the tools of modern simulation tools add-ins (such as Oracle Crystal Ball, Palisade @Risk, etc. ) in a simplified command line. Major features to cust modeling time and time to answer are discussed :
• Correlation of Simulated Inputs (Iman-Conover[1982])
o Covariance Matrix
o Rank Order Correlation Matrix
o Pearson Product Moment Correlation Matrix
• Data Analytics tools for analyzing both simulation results and exploratory analysis of historical datasets
o Density & Histogram Charts with Descriptive Stats
o Density
o Histogram
o TimeSeries Trend Chart
o Sensitivity Analysis Charts
• Sensitivity Analysis Chart
o Rank Correlation
o PPMC
o Contribution to Variance %
• TimeSeries simulation
o Time Series using Simulated Random Walk
o Time Series using Historical Data to calculate parameters for Simulated Random Walk
o Trend Charts with Customizable Confidence Bands
• Import / Export of results using the SIPMath standard.
I am currently the executive partner and analytics practice lead at Technology Partnerz Ltd., a firm specialized in delivering analytics solutions focused on forecasting, simulation and optimization to organizations all over the world. As practice lead, I have advised senior leadership on analytics strategy and implementation, trained hundreds of business analysts on predictive analytics, lead or supported risk analysis initiatives in finance, operations as well as project cost and schedule risk.