On the Semi-Optimization of Normal Probability
Yeah, this is normal alright.
For three stages, calculate the probability of the event on the top display, and enter it into the module.
Each event is an inequality involving a variable 𝐙. 𝐙 is a random variable (with a Normal frequency density, mean 0 and variance 1). The function tabulated on the next page is 𝚽(𝐳), which is the probability that 𝐙 < 𝐳.
Using the Table
If 𝐳 has one or two decimal places:
𝚽(𝐳) = the intersection between the ones / tenths and hundredths digits, on the left side of the table. Ignore the (ADD) column.
If 𝐳 has three decimal places:
Find the intersection between the ones / tenths and hundredths digits, on the left side of the table.
Also find the intersection between the ones / tenths and thousandths digits in the (ADD) column, and multiply it by 0.0001.
𝚽(𝐳) = the sum of these two numbers.
Some events in later stages may have greater-than signs, or negative values of 𝐳, etc. A list of these cases are found here:
- 𝐙 > 𝐳 OR 𝐙 < -𝐳: The probability is 1.0000 - 𝚽(𝐳).
- 𝐙 > -𝐳: The probability is 𝚽(𝐳). Treat this like 𝐙 < 𝐳.
- 𝐳1 < 𝐙 < 𝐳2: You will need to calculate for two probabilities in this case. Start with calculating 𝐙 < 𝐳2 (𝚽(𝐳2)), then 𝐙 < 𝐳1 (𝚽(𝐳1)), using the above cases if necessary. Subtract the result of the second from the first to get this probability.
TLDR: 𝐳1 < 𝐙 < 𝐳2 = 𝚽(𝐳2) - 𝚽(𝐳1)
Do not use a calculator for this module. Answers given by the table may be slightly different to those found with a calculator, and the module will only accept the former.