Coronavirus Pandemic – How the US Reached 300K Death Toll

(Click on the image to enlarge)

At the outset of the onslaught, when the White House Task Force projected a high death toll of 65K, the people worldwide woke up in fear and awe. Fast forward nine months: Yesterday, the widely-followed Johns Hopkins' Coronavirus Resource site flashed that the pandemic-related confirmed death toll in the US had crossed 300K. 

Of course, we have much to cheer on today as a couple of vaccines are ready (or being readied) for mass distribution, starting with the population's high-risk segments.

Source: https://www.worldometers.info/coronavirus/

 

Unfortunately, the actual death toll eclipsed the dire predictions in no time, hitting 100K on May 22, i.e., within three months from the first fatality, forcing the public health researchers to revert to the drawing board to tweak their vastly-flawed models that under-predicted, to say the least. New York and New Jersey collectively contributed roughly 43% of the overall death toll.

  

Source: https://www.worldometers.info/coronavirus/

The rate of growth in the death toll slightly eased while reaching 200K on September 15, thanks to the newly-approved treatment drugs like Gilead's Remdesivir and Regeneron's Regn-Cov2, replacing the malaria drug Hydrochloroqueen that was initially used by many hospitals to treat Covid patients. During this phase, while the Northeast had eased, the Sunbelt caught on with a fury.

Source: https://www.worldometers.info/coronavirus/

The above graphs demonstrate how the death rate lately has been accelerating, with a backward-bending tilt. Like the way we reached the first 100K deaths in just three months, we achieved the same distinction during the recent phase as we reached the 300K mark, despite all of the advancements in treatments and logistics.   

Now that the vaccines are here, hopefully the next milestone will be unremarkable.

Stay safe!

Data Sources:

https://www.worldometers.info/coronavirus/

https://coronavirus.jhu.edu/map.html

https://en.wikipedia.org/wiki/COVID-19_pandemic_in_the_United_States

-Sid Som

homequant@gmail.com

Coronavirus Outbreak – Recounting New York City vs. Rest of New York State

(Click on the image to enlarge)

By now, it's an open secret that the City of New York was the epicenter of the outbreak in the US in April. However, what most people, especially those in other states, do not know is that the rest of New York State ("State") was as badly impacted as the City of New York ("City").  

What everyone needs to know...

1. At the height of the outbreak in April, the rate of growth in total cases in the State exceeded that of the City. While the caseloads in the State grew 242% (from 40,572 to 138,624), the City experienced a slightly lower growth of 228% (from 51,809 to 169,690).  

2. The growth in May significantly tapered off for both the State and the City to 19% and 18%, respectively. The scatter graph demonstrates the classic "flattening of the curve" when the angle declined sharply from a linear 45 degrees to a curvilinear 30 degrees.

3. The taper has been continuing in June, with the growth declining to a mere 2% for both (in the first ten days of the month). The death tolls have also dropped off along with the total cases, thanks largely to the increased testing of the symptomatic candidates.

4. The regression graph shows that the outbreak was quite symmetric across the State, contrary to what the media led people to believe (that the City was the hardest-hit in the State). The reason the City experienced more cases is the higher population density than the rest of the State.

5. Since the pandemic affected the minority communities at a disproportionately higher rate, the City -- comprising a large minority population -- had a much larger death toll than the State. Out of the statewide death toll of 30,680 thus far, the City accounts for 17,292 and the remaining 13,388 goes to the State, revealing a very uneven split of 56% and 44%.

The good news is that the entire state is out of the woods, with no sign of re-emergence on the horizon whatsoever!

Stay safe!

Data Sources:
https://www.worldometers.info/coronavirus/
https://en.wikipedia.org/wiki/COVID-19_pandemic_in_New_York_City 

-Sid Som
homequant@gmail.com

Coronavirus Outbreak – How the Midwestern States are Faring!

(Click on the image to enlarge)

Highlights...

1. Though Michigan experienced an early outbreak, the general surge in the Midwest has been more recent. Therefore, the state-wise case totals are far lower than their East coast counterparts. For instance, Illinois, the most populous state in the region, has a current caseload of 83K vis-a-vis NY and NJ's 349K and 142K, respectively.      

2. Since the early outbreak was in Michigan, its death tolls are the highest among the Midwestern states. In fact, its death rate of 9.7% is currently the highest in the nation. Expectedly, its normalized (per million population) death rate is also the highest in the region. In that respect, Ohio, the second most populous state in the region, has been doing better than Illinois and Michigan.

3. The reason Ohio has done better thus far results from the lower positive rate of 12%, as compared to Illinois' 17.6% and Michigan's 15.1%. Even Indiana has a very high positivity of 16.7%. The polarized positive rates of Nebraska (17.9%) and North Dakota (3.3%) are, however, baffling, although they have managed to maintain low death rates.

4. In line with the high positivity, IL, IN and NE have high active caseloads as well. Minnesota (MN) and Dakota's have done an excellent job in managing their caseloads. MI has been proactive as well. 

5. In terms of the total tests administered, North Dakota with a normalized rate of 62K exceeds the national average of 30K. Ohio, Missouri and Kansas are way behind the curve with an average of 18K/each. In order to keep death rates and caseloads low, all of the Midwestern states should be targeting North Dakota's proactive testing average of 60K as the early detection is the key, until a vaccine becomes a reality.

Stay safe!

Data Source: https://www.worldometers.info/coronavirus/

-Sid Som
homequant@gmail.com

Coronavirus Outbreak – Is it Gender-neutral?

(Click on the image to enlarge)

There has been a lot of talk about the gender and age distribution of the coronavirus cases and deaths in the hardest-hit countries in Europe. Since Spain provides a large enough sample representing those countries, especially Belgium, France, Ireland, Italy, Netherlands, Portugal, and the UK, it is being used here to make the empirical case. This sample represents all confirmed cases through 05-04-20.

Overall, while there were 26,819 more cases for women, the death rate for them was significantly lower at 5.94% versus 10.46% for men.

1. Between the ages of 0 and 19, the infection and death rates for both genders were negligible (<1%). Considering that only the symptomatic cases were confirmed and closed, one can easily infer almost herd immunity for the asymptomatic candidates in those two age groups. 

2. In the next two age groups, i.e., 20 to 29 and 30 to 39, while the infection rates were much higher, the death rates were nonetheless negligible (<1%). One can, therefore, infer that these two age groups, irrespective of the gender, are prone to "mild to moderate" infections, thus easily leading to recovery.

3. In the next two age groups, i.e., 40 to 49 and 50 to 59, while the rates were expectedly higher, there were however no significant gender-wise differences in either rate. Simply put, between the ages of 0 and 59, gender was more or less insignificant, either way.

4. Gender however played a significant role in the next two age groups, i.e., 60 to 69 and 70 to 79. Both rates were significantly higher in men than in women, especially the death rate in the higher age group.

5. Then again, the rates were very similar in the next age group of 80 to 89, except that the death rate was a bit higher for women than for men. 

6. Gender again played a significant role in the 90+ age group, when both rates literally doubled for women.  

Stay safe!

Data Source: https://en.wikipedia.org/wiki/COVID-19_pandemic_in_Spain

-Sid Som
homequant@gmail.com

Coronavirus Outbreak – It’s Time to Overlay the Moving Average

(Click on the image to enlarge)

Now that the coronavirus outbreak has been in full swing for over six weeks, it's a good practice to study the daily data points alongside a simple trendline representing the 3-day moving average. This smoothes out the daily reporting anomalies, especially the the weekends and holidays. As the liquidity increases, the 3-day average can be extended to a 5-day and 7-day averages. 

This is very similar to daily closing vs. weekly closing prices for stocks and indices. When an extended series is available, the weekly closings provide a smoothed out trend. 

Likewise, if the current outbreak continues to plague us for an extended period of time, researchers would be looking at weekly trends rather than the daily trends. When a particular short stretch has to be looked at more closely, say between 5/1 and 5/15, obviously the daily data points remain more critical.

The advantage of a moving average trendline is that it, unlike the linear regression line, does not suppress the seasonal peaks and troughs. For instance, the moving average trendline of an extended two year outbreak, as in the case of the SARS outbreak, will still show the bi-modal allergy seasons (e.g., Apr-May and Oct-Nov), thus allowing researchers the opportunity to address or flatten those seasonal peaks.    

The above graphic shows that while NYS peaked around 4/10, there were follow-up spikes on 4/15 and 4/25. The 3-day moving average trendline, however, has moderated them down with two lower low peaks. Overlaying the moving average trendline is useful for presentations as well. 

When a forecasting model is needed out of an extended time series, the Autoregressive Integrated Moving Average (ARIMA) model is very handy. Hopefully, we will not reach that point as far as this outbreak is concerned.

Stay safe!

Data Source: https://en.wikipedia.org/wiki/COVID-19_pandemic_in_New_York_(state)


-Sid Som
homequant@gmail.com
       

Coronavirus – Is New Jersey Facing a Follow-up Outbreak?

New Jersey is not out of the woods yet. 

Unfortunately, in the last five days, the daily counts of confirmed infections and deaths have sharply trended up, proving that the early signs of a peak, as well as a taper, were false. 

The regression line (top graph) demonstrates how the linear trend turned into a logarithmic trend by tilting over at the intersection of 6K deaths and 105K infections. However, the most recent data points (4/28 to 5/2) not only flipped to the upside but also switched to the other side of the line, indicating the reversal of the prior trend. 

The bottom graph is more telling. When the data points were pared down to the most recent days -- between 4/21 and 5/2 -- the graph clearly demonstrates an emerging exponential slope, meaning sharply tilting back up. More precisely, the three most recent data points have not only reversed course, but have jumped back above the trend line as well.

The fact that NJ's death toll has soared 22 times (from 355 to 7,752) in just 32 days is truly alarming. Total infections have also climbed six-fold during the same period. 

Of course, whether this is a short-term trend reversal (due to some changes in the data capture methodology, i.e., non-statistical) or a follow-up outbreak remains to be seen. 

Given this sudden trend reversal, the Governor of New Jersey must not relax the current social restrictions until a more stable sideways trend, if not a downward trend with steady decline, emerges.  

Stay safe!

Data Source:
https://en.wikipedia.org/wiki/2020_coronavirus_pandemic_in_New_Jersey


-Sid Som
homequant@gmail.co

Comparing Coronavirus Stats – New York City vs. Rest of NYS

(Click on the image to enlarge)

The line graph (top) shows while the New York City curve has just started to flatten, the rest of the NYS curve has been flattening for a while (since 4/11). 

New York City is the home to a large minority population, especially African Americans and Latino Americans. This rapidly increasing divergence between the two curves results from the fact that the city's minority population has been more severely impacted by this outbreak than their counterparts. 

On 3/18 (not shown), the split between NYC and the rest of NYS was 644 to 730 (difference of -86 cases), growing to 1,339 to 1,043 (+296) the next day. The same trend has been continuing ever since; for instance, the difference grew to 10,483 on 3/31 and accelerated to 13,071 and 16,158 on 4/5 and 4/11, respectively. The difference has however tapered in the last three days, indicating the simultaneous flattening of curves.

By studying the two data lines mutually exclusively, one can easily come to the conclusion that the rest of NYS had peaked in early April, while NYC has just started to peak. By early next week, both curves are expected to show clear logarithmic slopes, meaning they will further tilt down.

The regression graph (bottom) shows the rapid acceleration of NYC cases: between 80K and 115K, the data points were pushed above the regression line. Now that the acceleration has significantly tapered, the five most recent data points are residing below the line. 

My blog post on 4/9 showed how the devil was getting frightened. Today, the devil is totally cornered. 

Humans will always win the war with the devil!

Data Source: 
https://en.wikipedia.org/wiki/2020_coronavirus_pandemic_in_New_York_(state)

Stay safe!

-Sid Som
homequant@gmail.com