# Lec 10 | MIT 18.01 Single Variable Calculus, Fall 2007

**Channel:** MIT OpenCourseWare
**Source:** https://www.youtube.com/watch?v=eRCN3daFCmU
**Transcript page:** https://www.withtranscript.ai/video/eRCN3daFCmU

## Chapters

- 0:00 — Chapter 1 (0:00)
- 0:40 — Chapter 2 (0:40)
- 6:38 — Chapter 3 (6:38)
- 8:35 — Chapter 4 (8:35)
- 14:29 — Chapter 5 (14:29)
- 18:03 — Chapter 6 (18:03)
- 20:57 — Chapter 7 (20:57)
- 23:22 — Chapter 8 (23:22)
- 24:38 — Chapter 9 (24:38)
- 27:07 — Chapter 10 (27:07)
- 29:21 — Chapter 11 (29:21)
- 30:44 — Chapter 12 (30:44)
- 32:35 — Chapter 13 (32:35)
- 35:41 — Chapter 14 (35:41)
- 39:56 — Chapter 15 (39:56)
- 40:58 — Chapter 16 (40:58)
- 43:35 — Chapter 17 (43:35)
- 47:00 — Chapter 18 (47:00)
- 51:00 — Chapter 19 (51:00)

## Transcript

**[0:00]:** The following content is provided under a Creative Commons license. Your support will help MIT Open Courseware continue to offer high-quality educational resources for free. To make a donation or to view additional materials from hundreds of MIT courses, visit MIT Open Courseware at ocw.mit.edu.

**[0:21]:** All right. So, we're ready to begin lecture 10. And what I'm going to begin with is by finishing up some things from last time.

**[0:33]:** We'll talk about approximations.

**[0:40]:** And I want to fill in a number of comments. And get you a little bit more oriented in the in the point of view that I'm trying to express about approximations. So, first of all, I want to remind you of the actual applied example that I wrote down last time. So, that was this business here. There was something from special relativity. And the approximation that we used was the linear approximation with the minus 1/2 power. That comes out to be t times 1 plus a half v squared over c squared. Now, the point I want I want to reiterate why this is a useful way of thinking of things and why this is that this comes up in real life. Why this is maybe more important than everything that I've taught you about technically so far. So, first of all, what this is telling us is the change in t divided by t, if you do the arithmetic here and subtract t, that's that's using the change in t is t prime minus t here.

**[1:56]:** If you work that out, this is approximately the same as a half v squared over c squared. So, what is this saying? This is saying that if you have this satellite, which is going at speed v, and little c is the speed of light, then the change in the watch that down here on Earth relative to the time on the satellite is going to be proportional to this ratio here. So, physically this makes sense. This is time divided by time, and this is velocity squared divided by velocity squared. So, in each case, the units divide out. So, this is a dimensionless quantity, and this is a dimensionless quantity. And the only point here that we're trying to make is just this notion of proportionality. So, I want to write this down just as summary. So, the error fraction, if you like, which is sort of the number of significant dis- digits that we have in our measurement, is proportional

**[3:00]:** in this case to this quantity. It happens to be proportional to this this quantity here, and the factor is

**[3:14]:** happens to be a half. All right? So, these proportionality factors are are what we're looking for. They're rates of change, the rate of change of something with respect to something else. Now, on your homework, you have something rather similar to this. So, in in in in problem uh on part 2b, part two, problem one, there's the speed of a pitch, right? And the speed of the pitch is changing depending on how high the mound is. And the point here is that that's approximately proportional to the change in the height of the mound. In that problem, we had this delta H. That was the X variable in that problem. And what you're trying to figure out is what the constant of proportionality is. That's That's what you're aiming for in this problem. So, there's a linear relationship, approximately. To all intents and purposes, this is an equality because the lower order terms are unimportant for the problem.

**[4:14]:** Just as over here, this function is a little bit complicated, this function is a little more simple. For the purposes of this problem, they're the same. Because there's the the the errors are negligible for the particular problem that we're working on. So, we might as well work with the simpler relationship. And similarly over here, so you could you could do this with in this case with square roots, it's not so hard. Here with reciprocals of square roots, it's also not terribly hard to do it numerically. And the reason why we're not doing it numerically is that as I say, this is something that happens all across engineering. People are looking for these linear relationships between the change in some input and the change in the output. And if you don't make these simplifications, then when you get, say, a dozen of them together, you can't figure out what's going on.

**[5:09]:** In this case, the design of the satellite, it's very important. This speed actually isn't just one speed because it's the relative speed of you to the the the satellite. And you might be at a It depends on your angle of sight with the satellite what the speed is. So, it varies quite a bit. So, you really need this rule of thumb. Then there are all kinds of other considerations in this question like, for example, there's the fact that we're sitting on Earth and so we're rotating around in what's called a non-inertial frame. So, there's the question of that acceleration. There's the question that the gravity that I experience here on Earth is not the same as up at the satellite. And that also creates a difference in time as a result of general relativity. So all of these considerations come down to formulas which are this complicated or maybe a tiny bit more, not really that much.

**[5:58]:** And then people simplify them enormously to these very simple-minded rules. And then they keep track of what's going on. So so in order to design the system, you you you must make these simplifications, otherwise you can't even think about what's going on. I mean it this this comes up in everything in weather forecasting, economic forecasting, um figuring out whether there's going to be an asteroid that's going to hit the Earth. These every every single one of these things involves dozens of these considerations. Okay, there was a question that I saw here. Yes.

**[6:38]:** Uh yeah, so in in basically any problem where you have a derivative, the rate of change also depends on what the base point is. That's the question. You're saying, "Doesn't this delta V also depend?" I had a base point in that problem. I happened to decide that pitchers pitch on average about 90 mph. Whereas in fact, some pitchers pitch at 100 mph, some pitch at 80 mph, and of course they vary the speed of the pitch. And so this varies a little bit. In fact, that's sort of a second-order effect. It does change the constant of proportionality. It's a rate of change at a different base point, which we're considering fixed. Okay? In fact, if you if that's sort of a second-order effect, when you when you actually do the computations, what you discover is it doesn't make that much difference to the to the A.

**[7:22]:** And that's something that you get from experience, that that it it turns out that which things matter and which things don't. And yet again, that's exactly the same sort of consideration, but at the next order of what I'm talking about here is you have to have enough experience with numbers to know that if you take if you vary something a little bit, it's not going to change the answer that you're looking for very much. And that's exactly the the point that I'm making. So, I can't make them all at once, all such points. All right. So, so that's my my pitch for for understanding things from this point of view. Now, uh we're going to go on now to to quadratic approximations, which are a little more complicated.

**[8:13]:** So, we talked a little bit about this last time, but I didn't finish, so I want to finish that this this up. And the first thing that I should say is that you use these use these when um the linear approximation is not enough.

**[8:35]:** Okay? So, you know, that's something that you really need to get a little experience with. Uh in economics, uh I told you they use logarithms. So, sometimes they use log-linear functions, sometimes they use log-quadratic functions when the log-linear ones don't work. So, most modeling in economics is with log-quadratic functions. And if you made it any more complicated than that, it's useless and it's a mess and people don't do it. So, they stick with the quadratic ones, typically. All right. So, the basic formula here, and I'm going to take the base point to be zero is that f of x is approximately f of zero plus f prime of uh zero times x. That's the linear part plus this extra term, which is f double prime of zero divided by two, x squared. And this is supposed to work for x near zero. Okay? So, I've chosen the base point as simply as possible.

**[9:34]:** All right. So, here's more or less where we left off last time. and one thing that I said I was going to going to explain which I will now is why it's a half f double prime of zero. So we we need to know uh need to know that. So let's let's work that out here. First of all so I'm just going to do it by example. So if you like the answer is just well, what happens when you have a a a parabola? A parabola is a quadratic. It had better it's quadratic approximation had better be itself. It's got to be the best one so it's got to be itself. So this formula, if it's going to work, has to work on the but on the nose for quadratic functions. So let's take a look. If I differentiate I get b + 2cx. If I differentiate a second time I get 2c.

**[10:32]:** And now let's let's plug it in. Well, we can um recover What is it that we want to recover? We want to recover these numbers a, b, and c using the derivatives uh evaluated at zero. So uh So let's see. Uh it's pretty easy actually. f of zero is a. That's on the nose. If you plug in x = 0 here, these terms drop out and you get a. And now f prime of zero, whoops. That was wrong. So I wrote f prime but what I meant was f. All right, so f of zero is a. Let's back up. f of zero is a. So if I plug in x = 0 I get a. All right, now f prime of zero that's this next formula here. f prime of zero, I plug in zero here and I get b. That's also good and that's exactly what the linear approximation is supposed to be.

**[11:25]:** But now you notice f double prime is 2c. So to recover c, I better take half of it. And that's it. That's that's the reason. There's no chance that any other formula could work, and this one does. All right, so that's

**[11:48]:** the explanation for the formula. And now I remind you that I had a collection of basic formulas written on the board, and I want to uh just make sure we know all of them again. So, first of all, there was sin x is approximately x, cos x is approximately 1 - 1/2 x squared, and e to the x is approximately 1 + x + 1/2 x squared. So, those were three that I mentioned last time. And again, this is all for x near zero. All for x near zero only. These are wildly wrong uh far away, but near zero they're they're they're nice good quadratic approximations. Now, the other two approximations that I want to mention are the logarithm. And we use the base point shifted so so we can put it at x near zero. And this one, sorry, this is an approximately equal sign there.

**[12:55]:** Turns out to be x - 1/2 x squared. And the last one uh is 1 + x to the power r, which turns out to be 1 + rx + r * r - 1 / 2 x squared. Now, eventually uh your mind will converge on all of these, and you'll find them relatively easy to memorize. Uh but uh it'll it'll take some getting used to, and I'm not claiming that you should uh uh recognize them and understand them all now, but we're going to put a giant box around this.

**[13:39]:** Yeah. Yeah, so the question was, you get all of these if you use that equation. Yeah, that's exactly what I'm going to do. Okay. So, I already did it actually for these three last time, but I didn't do it yet for these two. All right? But I I will do it in a in about 2 minutes. Well, maybe 5 minutes. But first I I want to explain just just a few things uh uh about these, okay?

**[14:10]:** All right, they all follow from the from the basic formula. In fact, that one deserves a pink box, too, doesn't it? Now, it's pretty important. All right. Yeah. Maybe even some little sparkles.

**[14:29]:** All right. Okay, so that's that's pretty important. All right. Almost as important as the more basic one without without this term here. All right, so now let me uh let me just tell you a a little bit more about the significance.

**[14:54]:** Again, this is just a to reinforce something that we've already done, but it's closely related to what you're doing on your problem set. So, it's it's worth you your while to to to recall this. So, there's this expression that we were dealing with, and we talked about it in lecture, and we showed that this tends to tends to e as k goes to infinity. All right? So, that's what we showed in lecture. And the way that we did that was we took the logarithm and we wrote it as K times Sorry, the logarithm of 1 + 1 over K. All right? And then we evaluated the limit of this. Now, I I want to do this limit again using linear approximation to show you how easy it is if you just remember the linear approximation. And then we'll explain where the quadratic approximation comes in.

**[15:48]:** So, I claim that this is approximately equal to K times 1 over K. Now, why is that? Well, that's just this linear approximation. So, what did I use here? I used log of 1 + X is approximately X for X is equal to 1 over K. All right? That's what I used in this approximation here. And that's the linear approximation to the natural logarithm. And this number is relatively easy to evaluate. I know how to do it. It's equal to 1. All right? That's the same Well, so where does this work? This works where this thing is near zero, which is when K is going to infinity. This thing is working only when K is going to infinity. So, what it's really saying this approximation for me is it's really saying that as we go to infinity in K, this thing is going to 1.

**[16:54]:** All right? As K goes to infinity.

**[16:58]:** So, that's that's what it's saying. That's the substance there. And that's how we want to use it in in in many instances to just to evaluate limits. We also want to realize that it's nearby when K is pretty large, like 100 or something like that. Now, so that's the idea of the linear approximation. Now, if you want to get the rate of convergence here, so the rate of what's called convergence. So, convergence means how fast this is going towards that.

**[17:34]:** What I have to do is take the difference. I have to take log AK and I have to subtract one from it. And I know that this is going to zero and the question is how big is this?

**[17:47]:** We want it to be very small. And the answer we're going to get so the answer just uses the quadratic approximation.

**[18:03]:** So, if I just have a little bit more detail than this expression here, in other words, I have the next higher order term. This is like 1 over K, this is like 1 over K squared. Then I can understand how big the difference is between the expression that I've got and its limit. Right, so that's what's on your homework. Right, this is on your problem set.

**[18:31]:** Okay, so that is more or less an explanation for one of the things that quadratic approximations are good for and I'm going to give you one more illustration. One more illustration. And then we'll we'll actually check these formulas. Yeah, another question. I have a question about Okay, when I give you a question, do they specify whether I should use linear or quadratic approximation? The question Very good question here. Um when they, which in this case means maybe me, when I give you a question, uh does one specify whether you want to use a linear or a quadratic approximation. The answer is in in real life, when you're faced with a problem like this, where some satellite is orbiting and you want to know the effects of gravity or something like that, nobody is going to tell you anything. They're not even going to tell you whether a linear approximation is relevant or a quadratic or anything.

**[19:33]:** So, you're you're you're on your own. When I give you a question, at least for right now, I'm always going to tell you. But, as time goes on, I'd like you to get used to when it's enough to use get away with a linear approximation. And you should only use a quadratic approximation if somebody forces you to. You should always start with trying with the linear one, because the quadratic ones are much more complicated, as you'll see in this next example. Okay, so the example that I want to use is you're going to be stuck with it, because I'm asking for the quadratic. So, we're going to find the quadratic approximation

**[20:15]:** uh near for X near zero to what? Well, this is the same function that we used uh in the last lecture. I think this was it. E to the minus 3X 1 plus X to the minus a half. Okay? So, unfortunately, I stuck it in the uh in the wrong place to be able to fit this very long formula here. So, I'm going to I'm going to switch it. All right? I'm just going to write it here.

**[20:57]:** And we're going to just do the approximation. All right? So, we're going to say quadratic in parentheses. And we'll say X near zero. All right? So, that's what I want. So now, here's what I have to do. Well, I have to write in the quadratic approximation for e to the minus 3x, and I'm going to use this formula right here.

**[21:25]:** And so, that's 1 plus minus 3x plus minus 3x the quantity squared over 2. The other factor I'm going to have to use this formula down here because r is minus a half. And so, that's 1 minus a half x plus a half times minus a half times minus 3/2 x squared. All right? So, this is the This is the r term and this is the r minus 1 term.

**[22:10]:** And now, I'm going to do something which is the only good thing about quadratic approximations. They're messy, they're long, there's nothing particularly good about them, but there is one good thing about them. Which is that you always get to ignore the higher order terms. So, even though this looks like a very ugly multiplication I can do it in my head just watching it. All right? Because I get a 1 times 1, I'm forced with that term here. All right? And then, I get the cross terms which are linear, which is minus 3x minus a half x. We already did that when we calculated the linear approximation, so that's this times the 1 and this times that one. And now, I have three cross terms which are quadratic. So, one of them is these two linear terms multiplying, so that's plus 3/2 x squared. That's minus 3 times minus a half.

**[23:05]:** And then there's this term multiplying the one. That's plus 9/2 x squared. And then there's one last term which is this monster here multiplying one. And that is -3/8.

**[23:22]:** Okay? So, the great thing is we drop x cubed, x to the fourth, etc. terms.

**[23:37]:** Yeah. Okay. Well, so copy it down and you work it out as I'm doing it now, all right? So, what I did is I multiplied one by one. I'm using the distributive law here. That was this one. I multiplied this 3x by this one. That was that term. I multiplied this by this. That's that term. And then I multiplied this by this. In other words, two x terms. That gave me an x squared and a -3 * a -1/2. And I'm going to stop at that point because the point is it's just all the rest of the terms that come up. Now, the reason The only reason why it's easy is that I only have to go up to x squared terms. I don't have to do the the higher ones.

**[24:21]:** Another question way back here. Yeah, right there. That last term Is it supposed to be negative? Oh, thank So, somebody can check my arithmetic here. Go ahead.

**[24:38]:** Why do I get to drop all the higher order terms? So, that's because the situation where I'm going to apply this is a situation in which x is say 1/100. Okay? So, this here's about a hundredth. Here's something which is on the order of a hundredth. This is on the order of a hundredth squared. A hundred squared, all of these terms. Now, these cubic and and quartic terms are of the order of 1/100 cubed. That's 10 to the minus sixth. And I'm The point is that I'm not claiming that I have an exact answer and I'm going to drop things of that order of magnitude. All right? So, I'm saving everything up to uh four decimal places. I'm throwing away things which are uh six decimal places out. Does that Does that answer your question? Okay. All right. So, that's the situation and now you can combine the terms.

**[25:39]:** I mean, it's not very impressive here. This is equal to 1 minus 7/2 x maybe plus uh 51/8 x squared. If I'd made that If those minus signs hadn't canceled, I would have gotten the wrong answer here. Anyway, so So, this is a two here, sorry. 7/2. All right, this is the linear uh approximation that we got last time and here's the extra information that we got from this calculation. Which is this 51/8 term. Here. All right, you have to accept that there's a certain degree of complexity to this problem and the answer is sufficiently complicated. So, it can't be less arithmetic because we get this peculiar 51/8 there. All right? So, so you you One of the things to to realize is that these kinds of problems, because they involve many many terms, are always going to involve a little bit of complicated arithmetic.

**[26:43]:** Okay, last little bit. I did promise you that I was going to derive these two uh relations. As I said, I did the ones in the left column. So, let's carry that out. And as someone just pointed out, it all comes from this formula here. So, let's just check it.

**[27:07]:** So, we'll start with the log function. This is the function f and then f prime is 1 over 1 plus x and f double prime, so this is f prime, this is f double prime, is minus 1 over 1 plus x squared. And now I have to plug in x equals 0. So, at x equals 0, this is log of 1, which is 0. So, this is at x equals 0. I'm getting 0 here, I plug in 0 and I get 1. And here, I plug in 0 and I get -1. So, now I go and I look up at that formula, which is way in that upper corner there, and I see that the coefficient on the constant is 0, the coefficient of x is 1, and then the other coefficient, the very last one, is minus a half. So, this is the minus one here, and then in the formula, there's a two in the denominator.

**[28:05]:** So, it's half of whatever I get for the second derivative at zero. All right, so this is the approximation formula, which is way up in that corner there. All right, similarly, if I do it for 1 plus x to the power r, I have to differentiate that, I get r 1 plus x to the r minus 1, and then r times r minus 1 x plus 1 to the r minus 2. So, here are the derivatives. And so, if I evaluate them at x equals 0, I get 1, that's 1 to the r power is 1, and here I get r, 1 to the r minus 1 times r and here I plug in x equals 0 and I get r times r minus 1.

**[28:59]:** So again the pattern is right above it here. The 1 is there, the r is there and then instead of r times r minus 1, I have half that for the coefficient.

**[29:21]:** All right, so these are just examples. Obviously, if we had a more complicated function, we might carry this out, but as a practical matter, we try to stick with the ones in the pink box and just use algebra to get other formulas. Okay.

**[29:41]:** All right, so so I want to shift gears now and and uh treat the subject that was supposed to be this lecture and we're we're not quite caught up, but we will try to do our best to to do as much as we can today. So the next topic is curve sketching and so let's get started with that.

**[30:17]:** Okay, so now happily in this subject there's there are more pictures and it's a little bit more geometric and there's relatively little computation. So let's hope we can we can we can do this. So so I I want to So here we go. We'll start with curve sketching.

**[30:44]:** And the goal here

**[30:54]:** Oh, you know, I So, that's like liner the last time I was That's kind of That's kind of sketchy uh spelling, isn't it? Uh

**[31:09]:** Yeah, there's certain kinds of things which I can't spell, but All right. Sketching. All right, so here's our goal. Our goal is to um draw the graph of F using um uh F prime and F double prime um whether they're positive or negative. Okay? So, that's it. This is the the the the goal here. However, there is a big warning that I want to give you, and this is one

**[32:02]:** that unfortunately I I I now have to make you unlearn, especially those of you who've actually had a little bit of calculus before. I want to make you unlearn some of your instincts that you developed. So, this will be harder for those of you who've actually done this before. All right? But, for the rest of you, it will be relatively easy, which is don't abandon your precalculus skills

**[32:35]:** and common sense. Okay? So, there's a great deal of common sense in this, and it actually trumps some of the calculus. The calculus just fills in what you didn't quite know yet. So, I will try to illustrate this, and because we're running a bit late, I won't get to the some of the main punchlines until next lecture, but I I want you to do it. So, for now, I'm just going to tell you about the general principles, and in the process, I'm going to introduce the terminology. What it just the words that we need to use to describe what it is that we're doing. And there's also a certain amount of carelessness with that in many of the treatments that you'll see. And a lot of hastiness. So, that just be a little patient, and we will we will we will do this. All right.

**[33:29]:** So, the first principle is the following. If f' is positive, then f is increasing. That's a straightforward idea, and it's closely related to this tangent line approximation, or the linear approximation that I just did. You can just imagine, here's the tangent line, here's the function, and if the tangent line is pointing up, then the function is also going up, too. So, that's all that's going on here. Similarly, if f' is negative, then f is decreasing.

**[34:10]:** And that's the basic idea. Now, the second step is also fairly straightforward. It's just a second-order effect of the same type. If you have f'' is positive, then that means that f' is increasing.

**[34:33]:** That's the same principle applied one step up, right? Because if uh f double prime is positive, that means it's the derivative of f prime. So, so it's the same principle just repeated. And now I would just want to draw a picture of this. Okay? Here's a picture of it, I claim, and it looks like something's going down, and I did that on purpose. But there is something that's increasing here, which is the slope. is very steep negative here, and it's less steep negative over here. So, we have the slope, which is some negative number, say -4, and here it's -3. All right? So, it's increasing. It's getting less negative. And maybe eventually it'll curve up the other way. And this is a picture of what I'm what I'm talking about here. That's what it means to say that f prime is increasing. The slope is getting larger.

**[35:28]:** And uh the way to describe a curve like this is that it's concave. So, f is concave up.

**[35:41]:** And similarly, um f double prime negative is going to be the same thing as f concave or implies f concave down.

**[35:57]:** All right? So, those are the those are the ways in which derivatives will help us qualitatively to draw graphs. But as as I said before, we still have to use our a little bit of common sense when we draw the graphs. These are just the additional bits of help that we have from calculus in drawing pictures.

**[36:25]:** So, I'm going to go through uh uh one example

**[36:34]:** to introduce all the notations, and then eventually, so probably at the beginning of next time I'll give you a systematic strategy that's going to work when um what I'm describing now goes wrong or a little bit wrong. So, let's begin with uh a straightforward example. So, the first example that I'll give you is the function f of x is equal to 3x - x cubed. Just as I said, to be able to introduce all the notations. Now, if you differentiate it you get 3 - 3x squared. And I can factor that. Right? This is 3 * 1 - x * 1 + x. Okay? And so, I can decide whether the derivative is positive or negative easily enough. Namely, uh just staring at this I can see that when um -1 is less than x is less than 1, in that range there, both these numbers, both these factors are positive.

**[37:55]:** 1 - x is a positive number, and 1 + x is a positive number. So, in this range f prime of x is positive. So, this thing is So, f is increasing. And similarly in the other ranges, if x is very, very large, this becomes if it crosses 1, in fact, this becomes this factor becomes negative, and this one stays positive. So, when x is bigger than 1, we have that f prime of x is negative, and so f is decreasing. And the same thing goes for uh the other side. When it's less than -1, that also works this way because um when it's less than -1, this factor is positive, but the other one is negative. All right? So, in both of these cases, we get that it's decreasing. So, now here's the schematic picture of this function. Uh so here's -1, here's 1.

**[39:12]:** It's going to go down, up, down. All right? That's what it's doing. All right, maybe I'll just leave it alone like this. All right? That's what it looks like.

**[39:28]:** So, this is this is the kind of information we can get right off the bat, and you notice immediately that it's very important from the features of the function, the sort of key features of the function that we see here, are these two places. All right? Maybe I'll even mark them in a like this. And these things are turning points.

**[39:56]:** Okay? Now, so so what are they? Well, they're just they're just the points where the derivative changes sign. Where it's it's negative here, and it's positive there, so there it must be zero. So, we have a definition, and this is the most important definition in this subject, which is that um if uh uh f prime of x zero is equal to zero, we call

**[40:26]:** x zero a critical point.

**[40:33]:** Okay, the word turning point is not used just because in fact it doesn't have to turn around at those points, but certainly if it turns around, then this will happen. And we also have another notation, which is the number y zero, which is f of x zero, um, is called uh, a critical value.

**[40:58]:** All right. So, these are the these are the key numbers that we're going to have to work out in order to understand what the function looks like.

**[41:17]:** Okay. So, what I'm going to do is just plot them.

**[41:28]:** We're going to plot the critical points and the values. Well, we found the critical points relatively easily. I didn't write it down here, but it's pretty obvious. If you set f of x equal to zero, that implies that uh, 1 minus x times 1 plus x is zero, which implies that x is plus or minus 1. So, those are known as the critical points. And now, in order to get the critical values here, I have to plug in f of 1 for instance. The function is 3x minus x cubed, so this is 3 times 1 minus 1 cubed, which is 2. And f of minus 1, which is 3 times minus 1 minus minus one cubed which is minus two. And so I can plot the function here. So here is the point negative one and here's up here is two. So this is whoops.

**[42:27]:** Which one is it? Yeah, this is negative one so it's down here I guess. So it's negative one negative two. And then over here I have the point one two. All right. Now what information do I get from So I've I've now plotted two I claim very interesting points. What information do I get from this? The answer is I know something very nearby because I've already checked that the thing is coming down from the left and coming back up. And so it must be shaped like this over here. The tangent line is is is zero. It's going to be level there. And similarly over here it's going to do that. So this is what we know so far based on what we've computed.

**[43:22]:** Question.

**[43:35]:** The question is what happens if there's a sharp corner? And the answer is um calculus is it's not called a critical point. It's a something else. And it's a very important point too. All right? And we will be discussing those kinds of points. They're much more dramatic instances of that. That's part of what we're going to what we're going to say but I just want to save that, all right? We will we will be discussing that. Yeah, question.

**[44:08]:** Uh the question that was asked was how did I know at the critical point that it's concave uh down over here and concave up over here. Uh the answer is that I actually did not use the second derivative yet. What I used is another piece of information. I used the information that I derived over here that f prime is positive where f prime is positive and where it's negative. So, what I know is that the graph is going down to the left of minus one. It's going up to the right here. It's going up here and it's going down there. I did not use the second derivative. I used the first derivative. Okay, but I didn't just use the fact that there was a that there was a turning point here. I I would Well, so actually I was using the fact that it was a turning point.

**[44:58]:** I wasn't using the fact that it was it had a second derivative though. Okay, for now. You can also see it by the second derivative as well. All right? Now, so now the next thing that that I'd like to do I I need to finish off this graph and I just want to do it a little bit carefully here in the in the order that that is is reasonable. Now, you might happen to notice and there's nothing wrong with this. Uh so so let let let's even fill in a guess, okay? Um in order to fill in a guess though and have it be even vaguely right, I do have to notice that this thing cross this function crosses the origin. Right? The function f of x is 3x minus x cubed happens to also have the property that f of 0 is equal to 0.

**[45:50]:** As again, common sense, you're allowed to use your common sense. You're allowed to notice a value of the function and put it in. Okay? So, there's nothing wrong with that if you happen to have such a value. So, now we can guess what our function is going to look like. It's going to maybe come down like this, come up like this, and come down like this. That could be what it looks like. But, you know, another possibility is it sort of comes along here and goes out that way. Comes along here and goes out that way. Who knows? It happens, by the way, that it's an odd function, right? These are all odd powers. So, actually, it's symmetric on the right half and the left half, and crosses it at at zero. So, everything that we do on the right is going to be the same as what happens on the left.

**[46:34]:** That's another piece of common sense. You want to make use of that as much as possible whenever you're drawing anything. You don't want to throw out information. So, this function happens to be odd. Odd, and f of 0 is equal to 0. I'm considering those to be kinds of precalculus skills that I want you to use as much as you can.

**[47:00]:** So, now here's the first feature, which is unfortunately ignored uh in in in most discussions of functions. And it's strange because nowadays we have graphing things, and it's really the only part of the uh uh exercise that you couldn't do, at least on this relatively simple-minded level, um without a uh uh with a graphing calculator. And that is what I would call the ends of the problem. So, what happens off the screen is the question. And that basically is the theoretical part of the problem that you have to address. You can program this, you can draw all the pictures that you want, but what you won't see is what's off the screen. You need to know something to figure out what's off the screen. So, in this case, I'm talking about what's off the screen going to the right or going to the left.

**[48:01]:** Okay? So, let's check the ends. So, here, let's just take a look. We have the function f of x, which is uh sorry, 3x - x cubed. Again, this is a precalculus sort of thing. And we're imagining now, let's just do x goes to plus infinity. So, what happens here? Uh when x is gigantic, this term is is is uh completely negligible, right? And it just behaves like - x cubed, which goes to minus infinity as x goes to plus infinity. And similarly, f of x goes to um plus infinity if x goes to minus infinity. All right? Now, let me pull down this picture again and show you what piece of the information we've got. We now know that it's is heading up this way. It doesn't go like this. It goes up like that, and I'm going to put an arrow for it, and it's going down like this.

**[49:02]:** All right? Heading down to minus infinity as x goes out farther to the right, and going out to plus infinity as uh x goes farther to the left. All right? So, now there's there's hardly anything left of this function to to describe. There's there's really nothing left except maybe decoration. And we kind of like that decoration, so we will we will we will pay attention to it. And to do that, we'll have to check the second derivative. So, if we differentiate a second time, the first derivative was um remember, 3 - 3x squared. So, the second derivative

**[49:47]:** is - 6x. All right? So now we notice that f double prime of x is negative if x is positive and f double prime of x is positive if x is negative. And so in this part it's concave down and in this part it's concave up.

**[50:19]:** And now I'm going to switch the boards so that you'll and and draw it. And you see that it's it's it was begging to be this way. Right? It wasn't it was uh right so we'll we'll fill in the rest of it here maybe in a a nice color here. So this is the whole graph and this is the correct graph. It comes down in one swoop down here and comes up here and then it changes to concave down right at the origin. All right. So this point is of interest not only because it's the place where it crosses the axis, but it's also what's called an inflection point.

**[51:00]:** Inflection point, that's a point where because f double prime at that place is equal to zero. All right. So it's a place where the uh where the second derivative is zero. Those we also consider those to be interesting points. Now so let let me just make one closing remark here which is that all of these this information fits together and we're going to have much much harder examples of this where you'll actually have to think about what's going on. But there's a lot of stuff protecting you and functions will behave themselves and turn around appropriately. Anyway, we'll we'll talk about it next time.
