П
CHINA’S CORONAVIRUS FIGURES DON’T ADD UP
by Kevin Ryan
Chinese officials today changed how they diagnose the coronavirus in their country, and as a result the total number of cases reported so far grew by 13,000 in one day, up to nearly 60,000 total.
The dramatically revised totals added to growing evidence that the authoritarian country is not providing accurate data. Indeed anomalies have shown up in China’s coronavirus numbers even before the change in methodology.
Perhaps the most mysterious and troubling discovery about the Chinese data is that, prior to today, the number of new deaths each day appears to correspond to a simple mathematical formula. Quantitative-finance specialists ran a regression of the data and found a near-perfect r-squared of .9999, meaning 99.99% of variance is explained by the equation.
“This never happens with real data,” one analyst said.
The finance publication Barron’s re-created the regression analysis of total deaths caused by the virus, and found the same variance. Melody Goodman, associate professor of biostatistics at New York University’s School of Global Public Health, also reviewed the data.
“I have never in my years seen an r-squared of 0.99,” Goodman said. “As a statistician it makes me question the data.”
For context, Goodman said a really high r-squared for public health data would be 0.70. “Anything like 0.99 would make me think that someone is simulating data. It would mean you already know what is going to happen,” she said.
As for why China would manipulate its data, there are several possibilities.
The first is that the country doesn’t have the adequate infrastructure in place to be able to count and report accurate data. Using a formula to estimate infections is easier than actually tabulating the number of patients. And since China likely didn’t want to be seen as unprepared or incapable of responding to the outbreak, the way it was during the SARS epidemic, the government may have chosen to report estimated data rather than admit it could not make accurate counts.
The second possible reason would be to use artificially controlled data in order to prevent world agencies from knowing the actual economic impact of the virus. Economists use the reported number of cases to calculate how damaging the outbreak is to the Chinese economy, and investors and others rely on such data to make financial decisions. Keeping control of the numbers that are reported means the Chinese central authorities can present a better picture for investors.
Economists say there is a longstanding data measurement problem in China, irrespective of the coronavirus. Government-released figures never match privately calculated data in China, with economic numbers often inflated.
Either way, the questionable data makes forecasting the severity of the virus and economic hit in China and beyond that much harder.
SOURCES: https://www.barrons.com/articles/chinas-economic-data-have-always-raised-questions-its-coronavirus-numbers-do-too-51581622840
https://www.wsj.com/articles/sharp-rise-in-coronavirus-cases-raises-questions-about-outbreaks-peak-11581617783
https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200210-sitrep-21-ncov.pdf
https://www.barrons.com/articles/investors-arent-buying-chinas-fairy-tales-1437019422
by Kevin Ryan
Chinese officials today changed how they diagnose the coronavirus in their country, and as a result the total number of cases reported so far grew by 13,000 in one day, up to nearly 60,000 total.
The dramatically revised totals added to growing evidence that the authoritarian country is not providing accurate data. Indeed anomalies have shown up in China’s coronavirus numbers even before the change in methodology.
Perhaps the most mysterious and troubling discovery about the Chinese data is that, prior to today, the number of new deaths each day appears to correspond to a simple mathematical formula. Quantitative-finance specialists ran a regression of the data and found a near-perfect r-squared of .9999, meaning 99.99% of variance is explained by the equation.
“This never happens with real data,” one analyst said.
The finance publication Barron’s re-created the regression analysis of total deaths caused by the virus, and found the same variance. Melody Goodman, associate professor of biostatistics at New York University’s School of Global Public Health, also reviewed the data.
“I have never in my years seen an r-squared of 0.99,” Goodman said. “As a statistician it makes me question the data.”
For context, Goodman said a really high r-squared for public health data would be 0.70. “Anything like 0.99 would make me think that someone is simulating data. It would mean you already know what is going to happen,” she said.
As for why China would manipulate its data, there are several possibilities.
The first is that the country doesn’t have the adequate infrastructure in place to be able to count and report accurate data. Using a formula to estimate infections is easier than actually tabulating the number of patients. And since China likely didn’t want to be seen as unprepared or incapable of responding to the outbreak, the way it was during the SARS epidemic, the government may have chosen to report estimated data rather than admit it could not make accurate counts.
The second possible reason would be to use artificially controlled data in order to prevent world agencies from knowing the actual economic impact of the virus. Economists use the reported number of cases to calculate how damaging the outbreak is to the Chinese economy, and investors and others rely on such data to make financial decisions. Keeping control of the numbers that are reported means the Chinese central authorities can present a better picture for investors.
Economists say there is a longstanding data measurement problem in China, irrespective of the coronavirus. Government-released figures never match privately calculated data in China, with economic numbers often inflated.
Either way, the questionable data makes forecasting the severity of the virus and economic hit in China and beyond that much harder.
SOURCES: https://www.barrons.com/articles/chinas-economic-data-have-always-raised-questions-its-coronavirus-numbers-do-too-51581622840
https://www.wsj.com/articles/sharp-rise-in-coronavirus-cases-raises-questions-about-outbreaks-peak-11581617783
https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200210-sitrep-21-ncov.pdf
https://www.barrons.com/articles/investors-arent-buying-chinas-fairy-tales-1437019422
