It seems to me that Factor Analysis or Principal Component Analysis could be used to determine how best to lose weight, i.e., what factors to focus on to maximize the chances of success.

If this has already been done by people with more resources than I, what were the results? How many factors must we be concerned about? What are they?

Searching this I get 200K hits. Can anyone summarize?

TIA.

I love your question. I think I spent more time thinking about this than anything else than I've done in months, sadly. As someone who has studied and applied multi-factor experiments to manufacturing processes, I'd have to say the biggest weakness here is finding a) 30 or more consenting adults who are willing to follow a carefully designed experiment to the absolute letter and b) are willing to have enough blood drawn to identify all the unique factors that make us metabolically different (at least those that have been identified so far). Researching and designing the experiment would be a major grant proposal in itself!

Very cool question! I am currently participating in a major cancer prevention study. I know what variables they are following, but I can think of at least 100 more without even pausing that could be more confounding. We humans are such fickle creatures! I am very glad that no one has bred us for captivity in a lab.

That's quite an answer!

I did find

"Handbook of assessment methods for eating behaviors and weight-related problems"
measures theory research
2nd ed
Allison, Baskin

but I'm having some trouble getting it.

While I'm waiting for my book through inter-library loan, maybe I do have some advice.

Each day for a week eat 1/4 lb estimated dry weight of the same food, and weigh yourself at least in the morning and in the evening.

Graph the results and use a trendline calculator to extract your daily weight loss.
Try to account for the variation. After 213 days I am still baffled half the time, gaining when I should have lost and losing when I should have gained.

The daily variation you will see must be water weight and is the "noise." The "signal" is the average weight loss over time as calculated above and this project exhibits a very bad signal to noise ratio.
For me the daily loss is 0.05 lb and the daily variation can easily be 2 lbs, plus or minus. Holidays & vacations have me gaining 8 to 10 lbs.

A Harper's journalist lost 30 lbs in 17 days by fasting but I don't seem to have the will power in the face of the temptations of modern, civilized living.

When I get serious, I do weigh daily, but I plot a weekly average. I use a basic correlation to get an idea of how things are going, and keep a sidebar of the rate of change of the slope of that line. And I did use control limits too. It worked really well.

Still, weight is but one factor. I once attempted to put together a weighted-ranking matrix of what was important to me and where I wanted to take this "better living" project. I had items on there like fat%, flexibility measures, strength evaluations, body measurements. As one of our long-term maintainers (saef) says, "Weight is nothing more than an indicator of other things in life." In essence, it's not a direct measurement at all. It *might* be a symptom of eating too much, but it also flags changes in stress, strength, hydration, overall health, hormone fluctuations, bowel function ... ya know what I mean? We measure it because it's easy and with a halfway decent scale, somewhat repeatable. But what does it MEAN? Much more difficult. Over on the Maintainer's thread, we deal with a lot of stuff like this. That and lots of pet talk,

You are really into this.

When I had a job that required travel I weighed the most ever, 192 lbs in the mid 1990s, and now I'm down to 156 (this morning). 150 might not be too thin.

I'm doing this to drop my cholesterol and it's not working too well. I could stand to also reduce my BMI.

Thing is, I don't really know why I want to prolong my life.
Maybe I'm trying to do what is recommended in the book, The Rectangular Curve, and it would also be good to outlive my enemies and be in good health while I do it!

My wife likes me rounder and I do look more muscular and more pleasing when I'm heavier, but, hey, it's fat.

These legs of mine are beginning to look like an anatomy textbook but my upper body less so. If I don't look like Ahhnold sometime soon Hollywood will never call me!
I'm lucky. . .some of the guys my age look like their butt somehow migrated into a large potbelly.

BTW, we have an FTC lawyer friend from Michigan ("Petarsky") and we did visit that island in the North accessible by ferry or by snowmobile. And we watched the sun sink into the water along with a bunch of other Sun God Believers. . .

I'm within 0.2 lbs of my goal weight of 150. Now the job is to increase my muscle to fat ratio, I guess by weightlifting and using some protein supplement. And to keep my weight within one lb or so of 150.

It probably helped that my wife is out of town and so there were no meal time cues. The last few days I've lived on water and lost two lbs/day. If my stomach bothers me at bedtime a hot bath somehow helps.
When I got hungry, distraction also helped - some TV, some solving of difficult textbook problems, leaving the house, etc..

One time I got trembly so I had a teaspoon of sugar.

At my height if I were muscled like Ahhnold I would weigh 185 lbs.

She gets back this week.
At this weight and me not being 18 anymore she may not like what she sees.
I don't totally, either, but my cholesterol is too high.

Good luck, folks. . .

Thanks, Wobbly! Congrats on your success, and on finding that happy balance. Consider perusing "The Cholesterol Myth" if you get an opportunity, just for another side of the story. You might find a diet adjustment that works for you better than any pill they can push.

Thanks, I will. . .