Tutorvo
Menu
Courses Guides Pricing Log in Sign up
Data & Analytics Advanced

Causal Inference for Business: Difference-in-Differences, Matching & Natural Experiments

Tell cause from correlation in Excel: confounding and Simpson's paradox, randomised tests, difference-in-differences with LINEST, parallel-trend checks, nearest-neighbour matching, propensity-score weighting, regression discontinuity and instrumental variables, in 8 graded labs on one retailer's data.

8 lessons 8 labs 2.2 h Certificate

  • One-time payment
  • 14-day money-back guarantee
  • Secure checkout with Stripe

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.

  • Show that a coupon which looks harmful in the totals actually helps in every segment, and compute a segment-adjusted effect.
  • Estimate the effect of a new store layout from a randomised test, check balance, and see why a change score is more precise.
  • Estimate the effect of Sunday opening with a 2×2 difference-in-differences table and confirm it with a LINEST regression with an interaction term.
  • Test the key DiD assumption with the pre-period gap trend, a placebo DiD and a line chart, then estimate the effect month by month.
  • Pair each trained store with the most similar untrained store and estimate the training effect on the trained stores (ATT).
  • Turn given propensity scores into IPW weights and estimate the effect of the mobile app on annual spend.
  • Fit a line on each side of the 1,000-point Gold cutoff and measure the jump in next-year spend.
  • Use a randomised reminder e-mail as an instrument and estimate the effect of the app on spend for customers who adopted because of the reminder.

About this course

Did the campaign raise sales, or would they have risen anyway? This course gives you the methods to answer that kind of question. You start with confounding and Simpson's paradox, then estimate an average treatment effect from a randomised experiment. Difference-in-differences is built as a two-by-two table and as a regression, and you check parallel trends with pre-period slopes and a placebo test. Matching finds nearest-neighbour controls, and propensity scores feed inverse probability weighting. Regression discontinuity measures the jump at a loyalty status cutoff, and an instrumental variable lab applies the Wald estimator to a randomised nudge. For analysts, data scientists, economists and product managers.

Curriculum

Course content

8 lessons · 8 labs · 2.2 h · final exam

  1. Correlation vs Causation: Confounding & Simpson's ParadoxLab15 min4 checked tasks
    Unlocked when you buy the course
  2. The Gold Standard: Randomised Experiment & Average Treatment EffectLab15 min4 checked tasks
    Unlocked when you buy the course
  3. Difference-in-Differences: The 2×2 Table and the RegressionLab18 min4 checked tasks
    Unlocked when you buy the course
  4. Checking Parallel Trends: Pre-Period Slopes, a Placebo Test & a Trend ChartLab16 min4 checked tasks
    Unlocked when you buy the course
  5. Matching: Nearest-Neighbour Controls with MIN, MATCH & INDEXLab18 min4 checked tasks
    Unlocked when you buy the course
  6. Propensity Scores & Inverse Probability Weighting (IPW)Lab17 min4 checked tasks
    Unlocked when you buy the course
  7. Regression Discontinuity: The Jump at the Gold-Status CutoffLab17 min4 checked tasks
    Unlocked when you buy the course
  8. Instrumental Variables: The Wald Estimator from a Randomised NudgeLab16 min4 checked tasks
    Unlocked when you buy the course
  9. Final exam12 questions, 75% to pass, unlimited retries.

Certificate

Finish with a certificate you can verify

Complete every lesson and pass the final exam to receive a certificate of completion for this course, with a unique credential ID and a public verification page.

  • Names the course, your score and the date of issue
  • A verification link for your CV or LinkedIn profile
  • 12 questions, 75% to pass, unlimited retries.

A Tutorvo certificate confirms that you completed a course. It is not an official or professional qualification.

Illustration. Your certificate shows your own name, course and credential ID.

Before you buy

Questions about this course

Is this a subscription?

No. You pay $9.99 once for this course. There is no subscription and nothing renews automatically. Prices are in US dollars; taxes may be added where required.

What exactly do I get?

All lessons and labs of this course, the final exam and the certificate, including later updates to the course, for as long as Tutorvo offers it. Access is personal and tied to your account.

Can I get a refund?

Yes. If you are not satisfied, email hello@tutorvo.com within 14 days of buying and we refund the course in full via Stripe. The guarantee does not apply once a certificate for the course has been issued to you, or where there are signs of abuse.

When can I start?

Right after payment. Access starts immediately at your request; for consumers in the EU/EEA the statutory right of withdrawal therefore ends once access has begun, while our 14-day money-back guarantee applies regardless.

Do I need Excel or other software?

No. The labs run in a spreadsheet inside your browser on a laptop, tablet or phone. Function names and syntax follow Excel.

Is the certificate an official qualification?

No. It confirms that you completed all lessons of the course and passed its final exam. It is not an official, state-recognised or professional qualification. The name is taken from your profile when the certificate is issued.

Full details are in our Terms · hello@tutorvo.com

Browse all Data & Analytics courses →
$9.99one-time payment
Buy course