Skip to content

INTERPOL: ANALYSIS OF VARIANCE FOR PREDICTING PURPOSES.

October 23, 2017

To my great delight, a few members of our Sunday discussion group in my lovely community in Canada accepted my offer to describe how I make successful scientific predictions? Making correct predictions takes years to learn. Here are a few that I have made: I have predicted the election of Justin Trudeau as Prime Minister of Canada, I have predicted the election of Donald Trump as president of the United States, I have predicted the incredible success of Elon Musk IPO shares of Tesla, and on the “personal” level I have predicted the ability of the “Gorter Model” of cancer treatment to cure my cancer within a year, and many more successful predictions. I cannot describe for you how to make this kinds of serious predictions, that takes statistics and years to learn, but let’s have fun about the topic as a “game.’

First Statistics: I came from Israel in 1961. My goal was to get a PhD in psychology. I failed Statistics (got a C-). I took the course again at UCLA and passed (got a B-). In graduate school I took advanced statistics and – failed (got a C). I took it again and got a B). In the PhD program I took high advanced statistics and got a B. PhD is a research degree so the professor said “You need an A.” I took it again and got an A. He said, “You never give up, do you? I said, “I want to use statistics to make impossible predictions easier.” He said,”Try Las Vegas.” I took students to Las Vegas on field trips to test 2 hypotheses, 1) “Affiliation as an aid to predicting success in Wheel of Fortune gambling,” and 2) “Cluster statistics.” Some students did better than chance. My statistics teacher at USC told me his biggest secret: He said, ” A valid analysis of Variance will have up to 5 variable, if you have more forget it. The weight of some of the variables will have to be determined subjectively, forget it if you think that you did it objectively. And the last one, don’t reject your variable (System 2) if one of the 5 variables turn out to be God, spiritual, metaphysical or “far out” in some way.

State of Palestine as a member of Interpol: Analysis of Variance of behaviour  for prediction purposes:

In the book Thinking: Slow and Fast, by Dr. Daniel Kahneman, the Nobel price winner in economics in 2002 there is a section on “Decision Making” called System One and System two behaviours. If you want to understand this posting better, reading his book, it may get you back on track.

The 5 behavioral variables that we are dealing with here are derived from System Two, 1) Interpol, 2) Israel, 3) Palestine, 4) The USA, 5) God and metaphysics. The “weights” assigned to each behavioural variable are subjective (System One). Choose your single prediction now (write it down). Choose your “weights.” The closer your weights estimates are to the “truth” (No one knows the truth) the more valid will be your prediction, meaning it may happened! (statistics is always probable, never absolute). Here is a sample of a prediction: “Palestine as a member of Interpol will improve or damage the probability of peace between Israel and the Palestinians?” Now, choose your prediction on any topic. Here is a sample of “Weights.” Choose your  “Weights” based on your prediction. Here is a sample of random ‘weights.” Palestine as a member of Interpol will improve or damage the probability of peace between Israel and the Palestinians.’

Interpol behaviour = 0% improve. 10% damage.

Israel behaviour = 10% improve. 30% damage.

Palestinian behaviour  = 0% improve. 30% damage.

The USA behaviour  = 0% improve. 20% damage  

God and metaphysics = 0% improve. 0% damage

Prediction (remember, this is only a Sample): The prediction is that Palestinian membership in Interpol will improve the probability of peace by 10% and will cause 90% damage: The probability of peace is lowered by 80%! (Remember again, this in only a sample). Now, make a prediction about anything yourself and see what you get. Plug-in your variables and weights and see what happens!

Note: Don’t predict anything seriously. For that you may need at least 5 years training! Do it only for fun or as a game with friends.

Advertisements
No comments yet

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: