Climate Models are simply multi-variable linear regression models. What that means is that they try to find a “best-fit” line between a single dependent variable and multiple independent variables that minimize the error between the regression line and the actual value. The classic multiple variable linear regression is the model for weight loss. Weight loss is largely a function of exercise and caloric intake. Those two independent factors are likely to explain 90%+ of the variation of the dependent variable weight loss. The remaining 10% may be due to your H20 intake, music you listen to while exercising, and your age and sex.

The problem with these kinds of models is that you don’t know what factors are significant and what ones aren’t, and how significant they are. You simply don’t know if exercise is responsible for 60% or 30%, caloric intake 30% or 60%, H20 5% or other, age and sex 5% or other, or what the weights or coefficients should be assigned to each variable. Also, if you leave out significant variables, or manually assign fixed weights to certain variables, the results are meaningless.

While I don’t know for sure, my analysis of the IPCC Climate Model Results leads me to believe that they fix a heavy weight on CO2, and either fix lesser weights to the sun, clouds and ocean temperatures and use ground measurements are used over satellite temperatures as the dependent variable. CO2 has been increasing in a linear manner, and the ground measurements have been “adjusted” to make them more linear, conveniently improving the output of the models. Satellite measurements have not been “adjusted” to fit the CAGW theory.

The easiest way to develop a real climate model is to simply allow the computer to objectively build the model for you, removed from bias, subjectivity, and preconceived outcomes. That is easily done with a computer statistical procedure called “Stepwise.” One simply takes the available data sets for Solar Radiation, Ocean Temperatures, Albedo, CO2, etc etc etc as the independent variables, and satellite temperature data as the dependent variable. Run the procedure “Stepwise” and it will run countless models adding some factors, dropping some factors, and changing the coefficients on the factors until it discovers the “best-fit” model. I am 1,000% percent confident, that if the IPCC was forced to run “Stepwise” on a model that includes all their data sets and using the Satellite Temperature Data there is 0.00% chance that CO2 would be one of the most significant variables.

This is a very very very easy to expose the fraud that CAGW truly is, and it would be done objectively and removed from political and personal bias. President Trump should demand that “Stepwise” be run on all Tax-Payer Funded climate models, and the results of the “Stepwise” model should be compared to the results of the existing model. If the R-Square, or explanatory power, of the “Stepwise” model, is greater than the existing climate model, then you have the case for possible scientific fraud. It will at least force the slimate clientists to acknowledge that there are factors far far far more important to the climate than CO2. Slimate clientists have chosen the computer as their weapon in this war, so President Trump should fight back using the computer. Like the Frankenstein Monster, the creation can be used to turn on its creators. Computers won’t lie, slimate climatists will and do, and “Stepwise” can prove it. The very computer models the IPCC created to manufacture the CAGW Myth can be used to destroy it.

Note: Correlation does not prove causation, but a causation requires correlation. StepWise finds the best correlations, not causation. It would, however, produce reports that would show the sun and ocean temperatures are far more significant than CO2 with regards to temperature. Slimate clientists would then be forced to defend CO2 against factors that do have correlations that point to causation and have to explain why they are ignored.

Please Like, Share, Subscribe, Comment and Forward this to President Trump

Barf! 🤮

Stepwise

blowsas a selection method.Stopping Stepwise: Why stepwise and similar selection methods are bad, and what you should use

Why we hate stepwise regression

Also, I would assume you’re not a professional statistician. I would not presume that climate models are a linear multiple regression problem — if they were, it would have been settled a long time ago.

A Nonlinear autoregressive exogenous model would be closer to what it is, if you also throw in coefficients that are functions of other variables in the model, and non-constant variance. These are not simple models.

LikeLike

The models I’ve seen are largely linear regressions of CO2 and Temperature. That is why they keep “adjusting” data so temperature is more linear. CO2 is linear, temperature isn’t, so that is why the adjustments are proof this is a fraud. Yes, stepwise isn’t perfect, but it definitely gets you in the ball park, especially if all selected variables for the theory being tested. Anyway, it is infinitely better than having biased conflicted “scientists” choose what variable are important and their weighting. Right now everything is designed and adjusted to make CO2 the most significant variable. Stepwise would never do that, and you can be certain it is unbiased.

LikeLike