Price Elasticity: Estimating Demand Curves, Log-Log Regression & Optimal Price

What you'll build

What you will be able to do

Each lab ends with a concrete result that you build yourself and the checker confirms, cell by cell.

• Measure the price elasticity of five coffee products from their last price change, with the simple and the arc (midpoint) method, and the point elasticity on a fitted demand line.
• Fit a constant-elasticity demand curve to 20 weeks of House Blend sales: logarithms, the slope as the elasticity, a forecast at a planned price and a scatter chart with trendline.
• Show that a price-only regression overstates the elasticity of the Cold Brew 4-pack when price cuts come with promotions, and fix it with LINEST and two dummy variables.
• Measure how strongly a competitor's price moves your House Blend sales, from one event and from a two-price regression, and forecast the effect of a competitor price cut.
• Find the revenue- and the profit-maximising price of the House Blend on a linear demand line, confirm the markup rule (Lerner index = −1 ÷ e) and apply it to a constant-elasticity model.
• Turn the fitted constant-elasticity model into a price table with units, revenue and profit, pick the best price and chart revenue and profit against price.
• Estimate separate elasticities for online and store customers, and turn each into a channel price with the markup rule.
• Size an online A/B price test from an assumed elasticity, then read the result: conversion, profit per visitor and the p-value of the conversion drop.