NOTE: This StatQuest was supported by these awesome people who support StatQuest at the Double BAM level: Z. Rosenberg, S. Shah, P. Govindaraj, J. N., J. Horn, J. Wong, I. Galic, H-S. Ming, D. Greene, D. Schioberg, C. Yee Lee, C. Walker, C. Lopez, C. Harder, J. Alexander, N. Nguyen Thranh, L. Cisterna, J. Varghese, K. Manickam, N. Fleming, F. Prado, J. Malone-Lee

Hello Josh,

Many thanks for your great video!

I’m just a little bit confused, in the slide at 10:00, what’s the meaning of the X-axis? The time consuming between meeting people? Whether the E(X) should be: E(X) = 10 * 0.4 + 20 * (0.4 + 0.2) + 30 * (0.4 + 0.2 + 0.1) + 40 * (0.4 + 0.2 + 0.1 + 0.09) + 50 * (0.4 + 0.2 + 0.1 + 0.09 + 0.05) + …

Your kind guidance is much appreciated!

Best regards,

Xiaochi

The x-axis is time between meeting people. 40% of the people were met in the first 10 seconds, thus, this event is labeled “10 seconds”, so we have 10 * 0.4. 20% of the people arrived between 10 and 20 seconds, thus, this event is labeled “20 seconds”, and we have 20 * 0.2 etc.

Thank you very much, Josh. Now I’m more clear about it.

What made me confused before is this labelling method. In the video, each interval is labelled by its “right end”. However, if we use “left end” or “middle point” to label each interval, the E(X) would be different. And this is exactly the shortage in discrete approximation.

Hi Josh,

I think this is one of your best videos yet (and I’ve seen most of them!).

May have a question though. Will post separately.

Great job! Thank you very much Josh!