Simulate Outcomes with SciPy and NumPyThis practical course introduces Monte Carlo simulations and their use cases. Monte Carlo simulations are used to estimate a range of outcomes for uncertain events, and Python libraries such as SciPy and NumPy make creating your own simulations fast and easy!
Apply New Skills in a Principled SimulationAs you learn each step of creating a simulation, you’ll apply these skills by performing a principled Monte Carlo simulation on a dataset of diabetes patient outcomes and use the results of your simulation to understand how different variables impact diabetes progression.
Learn How to Assess and Improve Your SimulationsYou’ll review probability distributions and understand how to choose the proper distribution for use in your simulation, and you’ll discover the importance of input correlation and model sensitivity analysis. Finally, you’ll learn to communicate your simulation findings using the popular Seaborn visualization library.
Introduction to Monte Carlo SimulationsFree
What are Monte Carlo simulations and when are they useful? After covering these foundational questions, you’ll learn how to perform simple simulations such as estimating the value of pi. You’ll also learn about resampling, a special type of Monte Carlo Simulation.What is a Monte Carlo simulation?50 xpDeterministic simulation100 xpStochastic nature of Monte Carlo simulation100 xpThe Law of Large Numbers100 xpResampling as a special type of Monte Carlo simulation50 xpSampling with replacement100 xpVisualization of resampling results100 xpPermutation practice100 xpLeveraging Monte Carlo simulations50 xpUnderstanding Monte Carlo simulations100 xpPaired dice simulation100 xp
Foundations for Monte Carlo
Now that you can run your own simple simulations, you’re ready to explore real-world application of Monte Carlo simulations across various industries. Then, you’ll dive into the heart of what makes a good simulation work: sampling from the correct probability distribution. You’ll learn about probability distributions for discrete, continuous, and multivariate random variables.The Monte Carlo process50 xpSteps of Monte Carlo simulations50 xpWrong deterministic calculation100 xpWrong input distributions100 xpGenerating discrete random variables50 xpSampling from a discrete uniform distribution100 xpSampling from a geometric distribution100 xpBetting between Tom and Eva100 xpGenerating continuous random variables50 xpChanging the mean of normal distributions100 xpChange the standard deviation of normal distributions100 xpTwo independent normal distributions100 xpGenerating multivariate random variables50 xpMultinomial sampling100 xpExploring a multivariate normal distribution100 xpMultivariate normal sampling100 xp
Principled Monte Carlo Simulation
Once you’re comfortable with your choice of probability distribution, you’re ready to follow a principled Monte Carlo simulation workflow using a dataset of diabetes patient characteristics and outcomes. You will explore the data, perform a simulation, and generate summary statistics to communicate your simulation results.Explore the data50 xpExamining tc, ldl, and hdl100 xpExamining y, tc, and cdl100 xpWhy do we need to explore the data?50 xpChoosing probability distributions50 xpTry other candidate distributions100 xpUsing visualization to guess a distribution100 xpInputs with correlations50 xpComparing simulated and historical data100 xpThe relationship between correlation and covariance matrices100 xpSummary statistics50 xpWhy do we need simulations?100 xpEvaluating BMI outcomes100 xpEvaluating BMI and HDL outcomes100 xp
Model Checking and Results Interpretation
Discover how to evaluate your Monte Carlo models and communicate the results with easy-to-read visualizations in Seaborn. Finally, use sensitivity analysis to understand how changes to model inputs will impact your results, and practice this concept by simulating how business profits are impacted by changes to sales and inflation!Evaluating distribution choices50 xpIntuition of probability distributions100 xpEvaluating distribution fit for the ldl variable100 xpVisualizing simulation results50 xpExplore HDL and BMI results100 xpExploring with box plots100 xpSensitivity analysis50 xpSimulation of a profit problem100 xpCompany sensitivity analysis100 xpSensitivity analysis using hexbin plot100 xpCongratulations!50 xp
Izzy WeberSee More
Data Coach at iO-Sphere
Izzy is a Data Coach at iO-Sphere. She discovered a love for data during her seven years as an accounting professor at the University of Washington. She holds a masters degree in Taxation and is a Certified Public Accountant. Her passion is making learning technical topics fun.