It's not
about being liberal or conservative anymore y'all. That is a hype offered by the fascist whores who want to confuse the people with lies while they turn this country into an aristocratic police state. Some people will say anything to attain power and money. There is no such thing as the Liberal Media, but the Corporate media is very real.
Here are the model's updated 14 and 21 day predictions:
39,768 = 797×15.0558×(1+0.0893565)^14
72,399 = 797×15.0558×(1+0.0893565)^21
Potentially almost 72,400 by early May unless we continue bending the curve.
Just to say, the assumptions made in the above model, are as follows:
The a value accurately measures the number of known positive cases, ie. 1 out of 15. Thus multiplying the a value to 797 is attempting to include the unknown cases.
The b value consists of a rate of increase that is both exponential and based upon a consistent percentage rate. The model has been dropping the percentage over the last 10 days, however, the rate of growth is still exponential.
If for instance we assume that only 50% of cases are known (a = 2), and the percentage growth is 8% (1+b = 1.08), then the prediction over the next 3 weeks is ...
8023 = 797×2×(1+0.08)^21
Another calculation I am using is the Gaussian regression model based upon increases per day. I sum over the interval where the model predicts 0.5 per day (currently at x=191 -- about 4.8 months from now) and I get 10,289 cases.
Thursday, 9 April 2020 at 20h 59m 59s
Grifters gonna grift
Because in chaos, they can steal.
Thursday, 9 April 2020 at 19h 33m 43s
Uh-Oh
Click here for an article describing how 51 cases of recovered patients might have caught or reactivated Coronavirus AGAIN !!
At a briefing earlier this week, Jeong Eun-kyeong, director-general of the Korean Centers for Disease Control and Prevention, said that 51 people previously thought to have recovered completely tested positive again shortly after they got out of quarantine. Jeong said the virus may have been “reactivated” in some fashion since it was in the CDC’s judgment too early after their recovery for them to have been reinfected.
A great analysis of the science on Salt in the diet
Click here for an article in Paleoleap that indicates just how faulty the research was that beget the mantra of "low-salt" good, "high-salt" bad.
The first study in 1972, was based upon feeding rats the equivalent (by body weight) of 250 teaspoons of salt per day. The second study in 1988 was an inter-population study that included 4 outliers (out of 52 groups) which influenced the weak correlation from positive to negative, and
...even the Intersalt study itself admitted that without the four outliers, “both regression analyses showed no significant associations of sodium with median systolic pressure” ...
By 1998, studies were increasingly finding fewer and fewer benefits of salt reduction, even at a public health level undetectable to individual subjects. In August of that year, Gary Taubes wrote an article summarizing the controversy and explaining why the link between salt and blood pressure (if it exists) is so difficult to determine. First of all, blood pressure is regulated by a complex homeostatic system: sodium affects it, but so do potassium, calcium, caloric intake, sex, age, and race. This introduces numerous complications to any study claiming that salt intake alone is responsible for high blood pressure.
Second, the early studies such as INTERSALT are all “ecologic” studies, comparing members of different populations (Yanomami Indians compared to Finns compared to Vietnamese compared to Americans). These studies appear to show that societies with a low salt intake have lower blood pressure, but they can’t account for the results of intrapopulation studies, which compare individuals within a certain population (white middle-aged men in Toronto). Within more homogenous population groups, researchers could find no direct relationship between dietary salt and higher blood pressure. Taking out the confounding factors that plague ecologic studies seemed to also reduce or eliminate the correlation between salt intake and blood pressure.
Latest SF Coronavirus update from the SF Dept of Public Health
Jumped up 54 from yesterday. I waited a few days since my last update because these crude models that I am using aren't very predicative and can fluctuate based upon incoming data.
Click here if you want to read an excellent article in fivethirtyeight from late March on how difficult statistical modeling can be.
Anyway, in my exponential model the b value has been decreasing, and is now .0944122, which means the rate of increase has been decreasing and is now less than 9.5%. The a value has however shot up and is now 12.7004. Which implies that there are 12.7 times more positives than we have actually measured/tested. So there could be upwards of 8,585 (12.7 times 676) positive cases.
I am keeping data for San Francisco, Alabama, California, Florida, Kentucky, Mississippi, New York State, and Tennessee, for the sake of comparison. Here are the current b values:
Percentage of increase based upon current COVID-19 positive cases as of 8 March 2020
San Francisco
9.44%
California
12.24%
Alabama
12.84%
Florida
13.68%
Kentucky
13.47%
Mississippi
11.99%
New York State
12.3%
Tennessee
12.56%
Here are the model's updated 14 and 21 day predictions:
30,359 = 676×12.7004×(1+0.0944122)^14
57,090 = 676×12.7004×(1+0.0944122)^21
I am using an additional model, the sideways "S" curve. That's called a logistic model, or 1 over 1 plus the euler growth factor. It enables prediction to the plateau or "flattening out", and I am using it to compare the exponential to the linear (the dashed red line). So you might notice that this morning's SF Department of Public Health release of 54 more cases puts the trend above the linear and logistic but less than the exponential. The trend has been above the linear for the last 5 days.
The small bell curve on the bottom regresses the daily increases from the previous date (which was +54 today). That model gets down to plus 0.5 cases when x equals 75, or 32 days after today, which is 10 May 2020.
Monday, 6 April 2020 at 23h 21m 53s
Excellent discussion
From Dr. MarkAlain Dery, at Tulane Medical Center, on the Coronavirus.
Monday, 6 April 2020 at 22h 59m 18s
This is the Common Cold on Steroids
More communicable than the flu, and 10 to 15 (or more) times lethal.
At timestamp 45:20 Dr Gundry discusses the myths and misconceptions about the coronavirus.
Namely:
It is a virus from a bat. It is not a weaponized genetic virus created by a biomedical lab. This is a natural virus that has lived in bats for a very long time.
Getting this virus is not a death sentence. Currently, in the USA, the mortality rate is 1.4%
This is not like getting the flu. Coronavirus is more deadly, and more contagious.
There is no certainty whether this is a seasonal virus. Some nutritionists think the correlation might be more related to fluctuations in Vitamin D levels over the year due to levels of sunlight more than actual seasonal variation due to the virus. The cold virus is not a seasonal virus however.
The cold virus mutates rapidly, which is why there is no vaccine for the common cold. It looks like so far over the last 4 months the coronavirus mutates more slowly. If that is the case, then a vaccine can hopefully be developed. However, the flu vaccines are ongoing because the flu virus mutates. Long story short, it's too early to know for sure. We just don't know if people can catch this again, or if the rate of mutation will create regular future infections just like the common cold or flu virus.
Monday, 6 April 2020 at 19h 55m 11s
About science based decision making processes
Over-reacting to non-randomized studies based upon small sample sizes with elements of the study excluded from the analysis of the resulting data set?
Sounds like it's time to be Presidential and all that.
Click here for a video explaining evidenced-based decision making from Perry Wilson, M.D. at medscape.
Here is a great article at Statnews that dissects the limitations of the small-sample studies that are being touted as indicating "some benefit".
Keep in mind that there is a huge problem with testing if a certain drug shows benefits to people who have a disease, and that is how to measure the "severity', because people will improve regardless of what treatments are given. People survived the blood letting during the 1700's to remove "the vigors" from the body, but that didn't mean it was due to the treatment. So in any study using small samples, people can and do improve despite the treatments, which makes it difficult to say whether the improvement was enhanced by the treatment.
The scientific method is currently in the boxing ring with con-artists and a public relations political campaign. Who will win? Who should win?