For the following project I'm supposed to complete, I am not fully understanding what it is exactly asking for. Am I supposed to run a code using the first equation or the second? What is the difference between the first and second equation?

Logistic map is given as:

$$x_{n+1} = r\cdot x_n(1 - x_n) ,\qquad x_n \in [0,1] \tag{1}$$ This recursive equation has different behavior for different values of $r$. The purpose of this project is to experience the effect of $r$ on this dynamic. To do so, let’s consider the member of the following set for values of $r$:

$$R = \left\{r_i \mid r_i = \frac{10}{i}, 1≤i≤1000\right\} \tag2$$

Now, we have the recursive equation, and we can calculate this sequence for any given initial values. In this step, I am interested in $x_{10000}$ for 100 random initial values $x_0$. In other words, you have to choose 100 random initial values. Run your code to find $x_{10000}$ for each of these initial values.

up vote 1 down vote accepted

You are supposed to implement both sets. Here is the idea. If you choose a single value for $r$, say $r=1$, then you have what is called a dynamical system that takes a value $x_0$ and maps it to a new value $x_1$, which it maps to $x_2$, and so on. If you pick a starting value $x_0$ then you can follow the trajectory. Say let's pick $x_0 = 0.5$:

$$0.5 \to 0.25\to 0.1875 \to 0.1523\to 0.1291\to 0.1125\to\cdots\to 0.00009989\to\cdots$$

each value of $x$ is followed by $r\cdot x\cdot (1-x)$, where remember we selected $r=1$. It's pretty clear from looking at this that the $x$ values are heading toward zero. Do they do that for every possible starting value $x_0$? Your instructor suggests that you pick a hundred random $x_0$'s and look at the trajectory of each one: for each one calculate the corresponding $x_{10000}$ that you get by applying the transformation $x\to r\cdot x\cdot (1-x)$ 10,000 times, and see where you get. Maybe make a scatter plot comparing $x_0$ with $x_{10000}$ in each case.

Then pick a different $r$, say $r=4$, and do the same experiment. The behavior of the system is extremely different! Most values of $x_0$ do not head toward 0. The scatter plot of 100 different $x_0$ values and their corresponding $x_{10000}$'s will look totally different.

So each experiment involves choosing one value for $r$, and looking at the behavior of the $x\to r\cdot x\cdot (1-x)$ map for various $x_0$'s, and seeing what you can notice about it. When you understand that map, try one with a different $r$.


  • Thanks, this clears it up quite a bit. I'm going to try this out. Thanks for editing my bad formatting btw. – Mike Jun 5 '13 at 18:11

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.