You are invited to Log in or Register a free Frihost Account!

# Exponential weight loose

nisibdv
I was thinling to design a diet to loose weight in an exponential form. This would be wonderful because the exponential loose of weight would make anyone happy of his diet because of the rapid initial loose of weight. I passing time one would approach asymptotically the desired weight without never reaching it really. So it would prevent someone to begin gaining weight again. So at the end the only object would be to maintain a constant weight. The advantage of the exponential loose would be tjha one can increase pregresibvely the ingestion of food as time go.

I havent tried it myself yet, but I am going to do it. I have to loose 20 kilos. Whe I do it I will tell you about the results. Maybe one more year.

Good bye and lets make diet!
Bikerman
I think you are talking complete bollox.
I'm tempted to lock this right here and now, but I'll wait to see if anyone has anything positive to say.
In the meantime you might like to consider what 'exponential' means and why the concept, applied to weight loss, is quite ridiculous. I presume you are talking about inverse exponents here (ie starting big and getting exponentially smaller), in which case you are proposing an immediate large weight loss which, as any dietician will tell you, is both undesirable and potentially dangerous.
The only sensible way to loose weight is gradually, by better diet and increased exercise. There are already too many stupid diet plans around, and they do more harm than good. Adding another, which relies on principles that are both unsound and probably not practically possible, is profoundly unhelpful.
nisibdv
My idea is not ridiculus and I could prove that. lets suppose the following model:

P(t)=a+O*exp(-b*t)

where O is the initial over-weight, a is the target weight and b is a mean time duration of the diet. At time t=0, the weight P will be:

P(0)=a+O

that is, the target weight plus the over-weight. and at time t tending to infinity (oo) the weight will be the target one:

P(t->oo) -> a

I have the following thesis: if you eat the daily amount of food corresponding to a healthy man with weight "a", then your loose of weight will be governed by an exponential of the type described above (exponential with a negative exponent, of course, you would explode otherwise).

So, with a not so strict diet the man doing diet will percieve a notable loose of weight inittially and that will make him happy.

I dont understand what can be wrong with that, why you got angry by my interesting post. Maybe you are fat like me and tired to make diets, so in this case I recommend that one, that as you see can not harm anyone because I am proposing to eat as a norma "a" kilos man would do.

I have also some reason to think that this exponential law will hold: the velocity in the loose of weight will be proportional to the actual overweight. That is a very resonable suposition.

Also, making some sport you could make the velocity of loose "b" greater, and I suppose that the factor "b" is different for different people. It would be interesting to research on that.

So, with the big amount of posts written only to gain points and fill the ciberspace with unuseful bits, you should reconsider your point of view about judging posts.

Cheers
nisibdv
Bikerman
 nisibdv wrote: My idea is not ridiculus and I could prove that. lets suppose the following model: P(t)=a+O*exp(-b*t) where O is the initial over-weight, a is the target weight and b is a mean time duration of the diet. At time t=0, the weight P will be: P(0)=a+O that is, the target weight plus the over-weight. and at time t tending to infinity (oo) the weight will be the target one:

OK, let's apply this then.
Example - person is 30kg overweight with a target weight of 120kg and a diet duration of 10 weeks

Week 1 - weight = 120 + 30^(-10*1) = 120 (plus a VERY tiny amount)
so they have to loose all the weight in the first week - the other 9 weeks are statistically irrelevant....

As I said - nonsense.
Genesiz
Not only is this completely unrealistic as a way to lose weight, but also incredibly dangerous and potentially lethal. Loosing such a massive amount of weight in one go (from your post it seems you are talking about weeks as opposed to months), would leave you with an awful amount of excess skin, which after you reach your 'target weight' can be very hard to get rid of. Your posts seem like you were saying that after loosing this weight, you would suddenly look completely different, and all this excess skin would just disappear. This is not the case.

Your idea is reminiscent of a Vertical Banded Gastroplasty, more commonly known as a stomach staple. This is regarded by the medical community as a 'very serious and dangerous' operation, and is used only in the cases of people who are clinically obese. Your idea of exponential weight loss is similar to the way in which people who have had a VBG loose weight, except that you have no clearance from your doctor to attempt this, nor it seems have you truly thought through the entire procedure. 1 in 100 people die from a VBG within one year, and if you do attempt this your chances would be a lot higher. Your body requires a certain amount of food to survive, and you can't just drastically reduce this and hope to succeed. The best way to loose weight is to join a group and loose weight in small increments, not all at once.

I must implore you not to consider this. If you do want to loose weight, go and see your doctor, tell him about this idea and I'm sure he would agree with me, and recommend better ways for you to lose weight.
nisibdv
Bikerman: you should be more carefull in your analysis and in judging the ideas of other people. In your example application you wrote

 Quote: Week 1 - weight = 120 + 30^(-10*1) = 120 (plus a VERY tiny amount

I dont understand why you put in the exponent "-10*1" maybe 10 referts to 10 weeks? and what is the number 1 there, the factor b of my model? remember that the factor "b" is an experimental factor that depends on the persons characteistics.

It is obvious that you didnt understand my model and your example is nonsense. I will explain it to you again.

Supposse a person weighting 150 Kg and it begins a diet to weight 120 Kg, that defines an ooverweight of 30 Kg:

a=120
O=30.

We DON'T KNOW factor "b" in the exponent. I am proposing the following: Let's make that fat man to eat the daily amount of food corresponding to a HEALTHY 120Kg man (This is for you to understand Genesiz that this is no an
 Quote: incredibly dangerous and potentially lethal
, the problem is that you followed the erroneous example of bikerman that didnt grasp the idea, I hope you do). In that case, I propose with well funded arguments (see previous posts of mine) that the rate of weight loss will follow an exponential law as described before:

 Quote: P(t)=a+O*exp(-b*t)

Bikerman: the factor "b" will depend on the subject, its units are 1/[t] and of course will be a SMALL factor. Maybe it confused you when I said " b is a mean time duration of the diet" when I should have said "b is the inverse mean time duration of the diet" and that mean time is an experimental factor that you will get if the subject eats the amount of food corresponding to a HEALTHY constant weight man of 120 Kg. But it was obvious in the context of my previous post and you would have grasped it if you had read it carefully. Excuse me if I repeat this for you to understand the idea.

Well any question on my proposed model are wellcome and excuse me if I hurt your feellings. I will be grateful to you if you consider carefully the model.
SonLight
The merits of the proposed diet are unclear. I do not believe the idea is inherently dangerous, as the point is to reduce caloric intake to the amount needed to sustain a desired target weight. Many diets restrict calories far below that point, and there is some real danger in doing that.

What I question is whether the weight loss will occur according to the formula. It seems a reasonable first-try hypothesis, but weight in humans does not seem to respond in so simple a manner. In fact, there is often a large initial loss when starting a diet due to water loss. Unfortunately, that has little real benefit and completely obscures the underlying loss of fatty tissue, which is presumably the point. Excessive initial water loss is one of the greatest dangers of a "starvation" type of diet.
Bikerman
 nisibdv wrote: Well any question on my proposed model are wellcome and excuse me if I hurt your feellings. I will be grateful to you if you consider carefully the model.

I don't understand the logic at all. You say b is the duration and that it is an unknown? How, then, do you solve the equation?
Why not demonstrate with the figures supplied - a 150kg person who is 30kg overweight. Show me!

The way you described your formula is that b is the duration of the diet and t is the specific time under consideration. Therefore P(t) is the weight at any specific time. I simply assumed a 10 week diet (ie b=10) and solved the equation for the first unit of time (week (t) = 1)). If that is not what you meant then you need to redefine your equation. You specifically said
 Quote: where O is the initial over-weight, a is the target weight and b is a mean time duration of the diet.

I also don't actually understand what 'exp' means. What exponent? Do you mean O^(-b*t) (that is what I assumed)?

I also don't understand what you mean by
 Quote: I am proposing the following: Let's make that fat man to eat the daily amount of food corresponding to a HEALTHY 120Kg man

Where is the term for this in the formula? It also makes little sense. The amount of food eaten daily by a healthy 120kg man (why man, by the way?) will depend on the level of physical activity (ie the calories burned). The requirement could be anything from around 2000 calories, for a relatively sedentary person, to 7000 or more for an athlete. There is no term for calorie intake in your formula so how is this input?
Genesiz
I still stand by my original post that exponential weight loss is dangerous. Especially if you haven't first consulted your doctor or other health professional. There are a lot of reports about people who have gone on DIY diets, the results of which have been less than favourable.

If we take your example of a man of weight 150kg who wishes to lose 30kg. It is likely that the amount he losses in the first week will be greater than the amount he losses in the 2+ weeks, but how much will he lose. If he loses half of the weight that is left each week, in 4 weeks time he has lost 94% of the weight he wishes to lose. Losing such a massive amount of weight in such a short time would be dangerous to that man's health, as your body cannot lose such a large amount of weight in one go without professional care (I refer to the VBG example i used in my original post). You would have a lot of excess skin, which you do not seem to have considered, which people who I have talked have said it can be very depressing at times.

I would also like to put these questions to you:

- in your original post, you said that after the weight-loss the person would feel happy about their diet. What proof do you have. The people I have talked to have said that the excess skin can at times be very depressing and embarrassing.
- have you talked to a doctor about you idea
- have you tried other diets, or talked to people who have tried other diets for their opinions.

A final point I would like to add is that you should heed Bikerman's words from his first post. I have talked to a dietician and told him about your idea. He said that such a large weight loss in one go would be very risky, as the results have not been proven and the hypothesis seems unsound. He did say that there is the minutest possibility (I cannot stress how small he said the possibility would be) that it may work, but that as a dietician he would never suggest this to any of his patients because of the large risks it would have.
nisibdv

 Quote: You say b is the duration and that it is an unknown? How, then, do you solve the equation?

I aswered that in a previous post:
 Quote: Maybe it confused you when I said " b is a mean time duration of the diet" when I should have said "b is the inverse mean time duration of the diet" and that mean time is an experimental factor that you will get if the subject eats the amount of food corresponding to a HEALTHY constant weight man of 120 Kg.

 Quote: Why not demonstrate with the figures supplied - a 150kg person who is 30kg overweight. Show me!

I can not show you with your figures because, as I said before, b is an experimental factor.

 Quote: I simply assumed a 10 week diet (ie b=10) and solved the equation

You applied the formula in a wrong way, because b is the inverse mean time duration, as I said before. Also, b is an experimental factor that depends on the subject, so you dont know a priori what b would be for a man (or woman) that eats a constant daily amount of food corresponding to a healthy man (or woman) with the targeted specified weight. Obviously, that constant daily amount of food would be recommended by prescription by a doctor, who would follow the evolution.

As SonLight pointed out (he understood clearly the model and made useful comments)

 Quote: I do not believe the idea is inherently dangerous, as the point is to reduce caloric intake to the amount needed to sustain a desired target weight. Many diets restrict calories far below that point, and there is some real danger in doing that.

I agree completely with him that the diet I am proposing is far less dengerous than the typical diets.

Also I agree with sonlight that, due to the excess weight due to retention of liquids, the exponential behaviour I am proposing could be affected.

 Quote: I also don't actually understand what 'exp' means. What exponent? Do you mean O^(-b*t) (that is what I assumed)?

In physics, whe one refers to an exponential behaviour, it doesn't matter what the base of the exponent is. Suppose exp is the neperian number e, then:

exp(x)= e^(x)=base^(alpha*x),

with base arbitraty and alpha=log_(base)(e), i.e. the base-logaritm of e. So only it would be a multiplicative factor on b. What is important is the functional behaviour proposed, not the specific parameters.

 Quote: Where is the term for this in the formula? It also makes little sense. The amount of food eaten daily by a healthy 120kg man (why man, by the way?) will depend on the level of physical activity (ie the calories burned).

Of course, bikerman, that amount will depend on a lot of variables of the person, (man or woman) but you could consult a doctor so as to obtain that data. The atractivenes of the diet woul be to create a constant habit of eating different from other diets where you begion by eating too little and when the subject gets tired of that attacs with every little piece of food he (or she) finds in the refrigerator.

And of course, don't expect to find in the formula a term for every aspect of the life of the subject, that is imposible even for the most complex model. That is first order approach to the problem. I think it is better that a man (or woman) gets a healthy habit of a constant intake of food than to make exigent diets where you are hungry almost all day.
Kerschi