How can people continue working during a “lock down”?

dlochner · Apr 4, 2020

John Nantz said:
Would be interesting to put what you did into an actual formula.

In the link on Conditional Probability there is are several formulas. When it comes to math, I've always been pretty good at understanding the concept and pretty lousy at the computations and algorithms, as I would say to my HS math teachers, "What do you mean prove the theorem? Any fool could see its true." Or at least I thought words like that.

The simplest demonstration of conditional probability is calculating the probability of getting 2 consecutive heads in a coin toss.

The probability of getting heads on the first coin toss is ½ or its decimal equivalent .5. The coin will be either heads or tails. Independently, the probability of getting heads on the second coin toss is also .5.

The probability of getting 2 consecutive heads is .5 * .5 = .25. Or the probability of event A times the Probability of event B given event A.
It is easy to see in a table

1st Toss. 2nd Toss
H T
T H
H H
T T

Those are all the possible combinations of a coin toss involving 2 coins, of those tosses only 1 out of 4 results yields 2 consecutive heads (tails). Don't be afraid to try this at home to test the theory. Your results won't be as neat the table, however, over a sufficient number of trials (30 or more) the results should be roughly the same, 25% of the coin flip pairs will be consecutive heads.

The grocery store example is probably over simplified however it goes roughly like this:

A = the probability of randomly picking one of 300 people out of a population of 470,000
B= the probability of one of those people 300 people being in a particular grocery store
C= the probability of randomly coming in contact with one of the 300 in a store with 100 people

The formula is something like the Probability C = Probability A * Probability B given condition A. That's a simple version, in reality it is probably more complex because we need to calculate B.

Mathematically conditional probabilities are always lower than the probability of either event occurring because the probability of any event occurring is always less than 1 and when fractions are multiplied the result is always less than the either fraction. ½ * ½ = ¼, ½ * ¼ = ⅛ and so on.

My venture down this road is not to dismiss the CDC recommendations or to encourage the blatant disregard for social distancing, hand washing, etc. It is really more about reducing anxiety about going to the grocery store and transferring it to anxiety about math.

Seriously, while the risk of contracting C-19 in any one event is relatively low, the more events (social contacts) that occur the higher the risk, or the more likely it is that someone will contract C-19. And if this were just the common cold, it wouldn't be such a big deal, but it is not the common cold and for many it will cause them to be hospitalized with life threatening consequences.

So, stay safe and find projects to keep youself busy, like figuring out combinational probabilities and field testing them, so what's the probability of getting 100 consecutive heads?

Hunter216 · Apr 4, 2020

Will Gilmore said:
We just extracted our first real quantity of honey from our bees

Have you considered “renting” your bees to orchards - farmers? Move the hive, get paid, keep the honey, farmer’s crop is pollinated to a higher degree.

JamesG161 · Apr 4, 2020

dlochner said:
so what's the probability of getting 100 consecutive heads?

≈1.26 10^30

But the key all models is the input and assumptions.

This big missing one, which is soon coming, is full data from the field testing.
This is very subtle...
Those who died and then tested positive
Verus
Those who tested positive and then died.

Dr Brix said 2 days ago the model had only 3% of data inputted and missing the other 97% :facepalm:

This weekend is expected to have real data starting show in models.

IMHO Sunday will tell the model... Tale.
Jim...

dlochner · Apr 4, 2020

JamesG161 said:
≈1.26 10^30

But the key all models is the input and assumptions.

This big missing one, which is soon coming, is full data from the field testing.
This is very subtle...
Those who died and then tested positive
Verus
Those who tested positive and then died.

Dr Brix said 2 days ago the model had only 3% of data inputted and missing the other 97%
This weekend is expected to have real data starting show in models.

IMHO Sunday will tell the model... Tale.
Jim...

Shouldn't that be ~1.26 * 10 ^-30 since it will be a fraction of 1?

JamesG161 · Apr 4, 2020

dlochner said:
will be a fraction of 1

No.
I said it last so I win.

Daveinet · Apr 4, 2020

Statistics and probability is only as good as the source data. A couple of things stand out. In your earlier example, you talked about the likelihood of comming in contact with someone who has the virus. This makes a huge assumption that the number of people with the virus is established. Due to limited testing, this is unlikely. Some areas are testing everyone with symptoms, other areas are only testing those who are admitted to a hospital. When you consider that only 20% end up in the hospital, this introduces a huge variability into the statistic. The real number is somewhere between 100% tested and 20% tested. With that level of variability, it is impossible to come up with any reasonably meaningful statistic. Add to the fact, contracting the virus and being aware of it, also has a large variability. How many unknown carriers are there?
Maybe the only meaningful statistic would be to work your analysis backwards. We have some awareness of the rate of infection. From that, we could derive the likelihood of contact, but we still have the variability of known cases vs undocumented cases.

Will Gilmore · Apr 4, 2020

Daveinet said:
Statistics and probability is only as good as the source data.

It is also heavily reliant upon reason and interpretation in context.
For instance:
The number of patients who die of Alzheimer's and weren't smokers far out number the number of Alzheimer's deaths in the smokers group. Data is perfectly good, but the early thinking was that smoking has a positive effect for preventing Alzheimer's.

As it turns out, that was a misinterpretation of the data because more people die earlier of smoking related decreases, leaving the pool of Alzheimer's patients short of a representative sample set. This data turns out to be poor only because it was misinterpreted.
One can't just look as data and assume the statistical conclusions are good because the data is good. Before anything else, statistics is about thinking, not about math.

-Will (Dragonfly)

dlochner · Apr 4, 2020

Daveinet said:
Maybe the only meaningful statistic would be to work your analysis backwards. We have some awareness of the rate of infection. From that, we could derive the likelihood of contact, but we still have the variability of known cases vs undocumented cases.

I tried to accommodate that uncertainty. When I wrote that there were 301 known cases of C-19 out of a population of ~470,000. In my example, I used 3,000 cases assuming that only 1 in 10 who have the virus were identified. The numbers being reported change every day and it is not clear (at least in the data being reported locally) whether that 301 number includes people who have been diagnosed and then recovered or just active cases.

I'll let the statisticians and epidemiologists calculate more precise risk factors. The point I was trying to make, is with reasonable precautions individually we are at a lower risk level for any given event, i.e., going to the grocery store. Of course if the conditions are changed, more frequent, longer, less protected, and closer contact, that risk level changes.

The number of people reported as being infected is large, over a 1 million today. However, that number also needs to be placed in context, there are 7 or 8 billion people in the world, in context roughly 1 person in 7,000 has been confirmed as being infected.

The tough part is the uncertainty. People infected with C-19 don't wear a big Scarlet C on their clothing. In times of uncertainty risks can be exaggerated, emotional responses (and other factors) to risk can alter the perception of risk. Currently the perception of risk ranges from the "We're all gonna" chicken little level to others (like the spring break students) that it is no big deal and of little risk. The more objective and realistic the risk assessment is, the better informed we can be about our behavior and how to manage that risk.

While not mentioned in the linked article (and a quick internet search didn't reveal a good reference) psychologists have found that people often over estimate risk factors, an example is a fear of flying. A person is much more likely to be killed by a drunk driver or a gun than to die in an airplane crash. Yet, many people are more fearful of flying than walking across a street or driving. There are many factors that contribute to poor risk assessment, however, a rational assessment of true risk is not one of them.

Again, my intent in bringing this up is to help people make a rational assessment of the risk they face. When people behave rationally they can act in the common good, which is what the CDC guidelines encourage, act in ways to slow the virus spread so others don't get sick and hospitals are not overwhelmed. When risk is inaccurately assessed, especially by a large group, we get irrational behavior that is not in our common good. Witness the panic buying of toilet paper.

As always, be safe, be smart, follow CDC guidelines for hand washing, social distancing, and so forth.

Will Gilmore · Apr 4, 2020

Hunter216 said:
Have you considered “renting” your bees to orchards - farmers? Move the hive, get paid, keep the honey, farmer’s crop is pollinated to a higher degree.

Yes I have. There are a lot of beekeepers who haul whole flatbed trailers full of hives around to farmer's fields. We don't have many hives and moving hives is hard on them. We don't have this beekeeping thing down yet. Renting them out is probably not going to be a direction we go, however. Look up Slovenian Hives. That is something we are thinking about for the future.

-Will (Dragonfly)

jssailem · Apr 4, 2020

Here you go John.. His more recent observations.
How we must respond to the coronavirus pandemic

dlochner · Apr 4, 2020

John Nantz said:
Sorry guys, gotta keep this short because my significant other is complaining ... er, telling me, that I have to get my work done and quit dealing with this virus. So, here is a short one ... kinda "back to the future" ...
.... from April 3, 2015!
Do the math, that was 5 YEARS ago!

The next outbreak? We’re not ready / Bill Gates Apr 3, 2015

8:36
So, what went wrong? Was anybody listening???

No. And those few who were listening, weren't listened to. The failure of leadership is sinful.

The administration knew how ill prepared the country and world for a pandemic last summer.

Before Virus Outbreak, a Cascade of Warnings Went Unheeded (Published 2020)

Government exercises, including one last year, made clear that the U.S. was not ready for a pandemic like the coronavirus. But little was done.

www.nytimes.com

Hunter216 · Apr 4, 2020

Will Gilmore said:
Yes I have. There are a lot of beekeepers who haul whole flatbed trailers full of hives around to farmer's fields. We don't have many hives and moving hives is hard on them. We don't have this beekeeping thing down yet. Renting them out is probably not going to be a direction we go, however. Look up Slovenian Hives. That is something we are thinking about for the future.

-Will (Dragonfly)

Well that is a nice approach to the entire process. I had a little experience as a youth helping someone out with their hives, but it was all outside and frankly brutal..

If you haven’t seen this you might want to watch.

“The Biggest Little Farm” on Netflix

https://www.netflix.com/title/81031829?s=i&trkid=13752289

Daveinet · Apr 5, 2020

dlochner said:
No. And those few who were listening, weren't listened to. The failure of leadership is sinful.

The administration knew how ill prepared the country and world for a pandemic last summer.

Before Virus Outbreak, a Cascade of Warnings Went Unheeded (Published 2020)

Government exercises, including one last year, made clear that the U.S. was not ready for a pandemic like the coronavirus. But little was done.

www.nytimes.com

There are some problems with blanket statement like this. Hindsight is always 20/20. As it turns out, compared to our European counterparts, we are way more prepared. We have 5 to 8 times more hospital beds per capita, depending on which country you compare to. Way more respirators as well. So when someone decides where to spend money, you are looking at a lot of priorities. You also consider the odds of it occurring to know how fast you have to move. If nothing happened, and we spent all kinds of money on it, it would be considered wasteful. Government moves slow. It always has. Based on history, what were the real odds of another pandemic occurring and in what time frame? Even so, just like moving money in the stock market, you don't do everything all at once. Truthfully when you consider that the current administration has been forced to spend the last 3 years defending itself. Regardless of if the accusations are true or not, the fact that this has dominated politics for the last 3 years, its amazing that anything else got done.
Thinking that knowing we had a problem last summer doesn't solve anything. Solutions don't move that quickly. What kind of solutions would work? New York had it in their budget for making up their short fall several years back and yet did nothing about it. Fact is, the right kind of solutions take significant time and organization to implement. I work for a company that builds equipment used in making respirators and face masks. We are filling some huge orders right now, but here is the problem. The company who is buying them is counting on these machines being used, but once the pandemic is over, these machines go in storage till the next time they are needed. That won't work, even though everyone involved doesn't know that. You can't just pull equipment out that has been sitting for ten years and expect it to operate. All the lubrication in the seals dry out with non-use. Electrolytic capacitors need to have voltage on them to maintain their properties. The microprocessor controls all have batteries to maintain certain types of information. This is the nature of modern electronics. Not only that, but the electronic industry moves so fast, that the components that make up this equipment are obsolete in only a few years, meaning that support for 10 year old equipment is somewhere between difficult and impossible.
The only real solution is for private industry to partner with government to build equipment that can be repurposed to fill the needs when emergencies arise. That is difficult to implement, as with most projects, margins are so low that building machines with dual purpose is uneconomical. The repurposing that is being done right now is mostly being done at a huge cost and sacrifice to private industry. In other words, many companies are doing this at a loss, just to help out.

dlochner · Apr 5, 2020

Assigning blame, pointing fingers, looking for faults, and making excuses does little to unite us in dealing with the pandemic. As a country our preparation was poor and we got off to a slow start. There's plenty of blame to go around, the beltway pundits can have a field day. And we'll get to make some choices next November. In the meantime we need to contain and tame the virus.

This afternoon I listened to an interesting podcast about how we have responded to the pandemic and the kinds of cognition that drives our behavior. They explored how social interactions, cognitive biases, and neuroscience can explain our response and why there is no TP in the grocery store. It is a long podcast at about 2 hours but very much worth the listen. You can find the podcast here:

YANSS 177 – Why people waited so long to take precautions against COVID-19, how to better persuade those who still refuse, and how to take better care of your mental health during isolation

Flatten the curve. That idea has spread through the population faster than COVID-19 ever could. That’s the power of culture, of human psychology, of brains interacting with brains. Of course, cultu…

youarenotsosmart.com

JamesG161 · Apr 5, 2020

If you use the models shown to us, you must start the model at certain point.
Here is a timeline chart for C19

Many think that for the most of us, were are past the APEX.
The hot spots are not.
Jim...

PS: We are not anywhere close to overloaded and I live 3 counties away from New Orleans.

dlochner · Apr 5, 2020

A quick quiz.

The number of virus infections doubles every day for 60 days at which time everyone will be infected. On what day will half the population be infected?

Ralph Johnstone · Apr 5, 2020

dlochner said:
On what day will half the population be infected?

59th day

What do I win

? ? ? ?

JamesG161 · Apr 5, 2020

dlochner said:
The number of virus infections doubles every day for 60 days at which time everyone will be infected. On what day will half the population be infected?

Wrong approach. That is called Gamblers Ruin.
Use the Fibonacci Series for right answer.
Fibonacci found that was better predictor since some candidates died before they infected others.
_____
Also you were right about Probabilities. They are always less than one or 100%
I was computing ODDS. Gambler that I am.
Jim....

How can people continue working during a “lock down”?

.

.

.

SBO Weather and Forecasting Forum Jim & John

.

SBO Weather and Forecasting Forum Jim & John

.

.

.

.

.

SBO Weather and Forecasting Forum Jim & John

.

.

.

.

SBO Weather and Forecasting Forum Jim & John

.

.

SBO Weather and Forecasting Forum Jim & John