Monday , April 6 2020
Home / Tag Archives: 17. Making decisions with uncertainty

Tag Archives: 17. Making decisions with uncertainty

Sweden’s strategy, for the long haul

Lockdowns will slow the virus (temporarily?), but will damage the economy (temporarily?).  Sweden is trying a different strategy. “It is important to have a policy that can be sustained over a longer period, meaning staying home if you are sick, which is our message,” said Tegnell, who has received both threats and fan mail over the country’s handling of the crisis.“Locking people up at home won’t work in the longer term,” he said. “Sooner or later people are going to go out anyway.” Let's...

Read More »

Prediction market: Who will be the Democratic Nominee?

 Each contract pays off $1 if the event occurs, e.g., Joe Biden wins the Democratic Nomination. The price of the contract can be interpreted as the probability of the event, Price=probability*$1.  For example, the Joe Biden contract is trading at $0.39 suggests a 39% that he will win the nomination.  These kinds of prediction markets can be used by firms to, e.g., predict future sales or profitability.

Read More »

Privately funded, Randomized Control Trials for policy

Results from the first four RCT's funded by Arnold Ventures.Here are the results for charter schools: The study found that students who won a KIPP middle school admissions lottery were 6 percentage points more likely to enroll in a four-year college than students who lost the lottery (47% of lottery winners enrolled vs. 41% of lottery losers). We view this finding as highly suggestive but not yet strong evidence of an effect because it did not quite reach statistical significance (p=0.085)....

Read More »

3 Randomista’s win Nobel prize in economics

...for using Randomized Control Trials to figure out which policies work and which do not.  Here are some blog posts on information from randomized control trials: Before going on a date, consult an economist Randomistas: fighting poverty with science What do physicians have in common with pilots? Book recommendation: Randomistas Below is a Ted Talk from one of the winners (I suspect that it would be profitable for business to run more randomized control trials, e.g., to estimate the...

Read More »

Learn regression simply

]]> ·         PEDAGOGY: The program trialandstderr.com "inverts" the problem of teaching regression by asking students to create data by clicking on an (x, y) graph to achieve a given outcome, like a statistically significant regression line. An early version of this program was used to teach Justice Dept. attorneys enough about regression to allow them to cross-examine rival experts. ·         © 2019, Luke M. Froeb & Keyuan Jiang. The program may be freely used at...

Read More »

What do physicians have in common with pilots?

They both respond to performance metrics (hand-washing or physicians; fuel consumption for pilots)--even without a reward (or penalty)!PHYSICIANS: METHOD: The ICU unit coordinator was trained to observe and measure hand hygiene compliance. Data were collected on hand hygiene compliance at room entry and exit for 9 months. Percentage compliance for each medical and surgical subspecialty was reported to chiefs of service at the end of each month. Comparative rankings by service were widely...

Read More »

Why do women bid less for gigs?

They offer 4% lower prices, win jobs more frequently, and earn higher expected revenue (prob[win]*price) than men:  New working paper: ...we provide empirical evidence for a statistically significant 4% gender wage gap among workers, at the project level. We also find that female workers propose lower wage bills and are more likely to win the competition for contracts. This raises the obvious question, whether women bid more aggressively than men because they think that the opportunity cost...

Read More »

AI makes marketing more powerful

There are two kinds of Artificial Intelligence (AI): symbolic logic, where automated rules replace human decision making; and machine learning, where outcomes, like sales, are related to policy, like advertising copy, to infer causality.  This from the WSJ seems to describe machine learning by Persado software: In one test, a headline by human copywriters urged consumers to “Access cash from the equity in your home,” with the call to action “Take a look.” A variant created by Persado was...

Read More »

What’s the best way to provide opportunity?

Good profile of economist Raj Chetty who finds that housing vouchers which allow poor families to move to "opportunity" neighborhoods, help younger kids catch up to richer peers.  President Trump signed a bill that uses the research to guide policy: Tenants have just started moving, but the program is already successful: The majority of families who received assistance moved to high-opportunity areas, compared with one-fifth for the control group, which was not provided with the extra...

Read More »

Does decreasing Type II errors leads to more Type I errors?

TYPE I ERROR:  False prosecution TYPE II ERROR:  False non-prosecutionUntil 2011, Title IX (the law that prohibits sex discrimination at federally funded schools) was rarely enforced because of a strict standard of proof. That changed under President Obama.  With lower standards of proof and evidence rules that favored prosecution, we should see Type II errors fall, but with a corresponding increase in Type I errors. Defamation claims are the new legal tool for men to clear their name and...

Read More »