 Okay, we've run our Monte Carlo simulation and we produced some output. In this particular case it's the news boy again and we got an output there of the demand and profit for 1000 trials. And I added a column here just to help me visualize. I coded it using an if statement. If the value here is greater than 4, if h4 is greater than 4 print out positive and that's just using the word we want with the double quotation marks before and after it, else print out negative. And so that gives me this particular value and then I use the conditional formatting to color it green if it said positive and red if it said negative. So that's how I got that column. But how do we use this to make some decisions and that's the whole point. Can generate a bunch of statistics using the descriptive statistics tool in the data analysis if you want or just plugging in some individual values here. I've got the mean and standard deviation of the profit, the max and the min and then assuming that this was, the output is a distribution that is suitably modeled by a t distribution, I found the confidence interval using the confidence dot t which requires the alpha 95% would be 0.05 and then the standard deviation there of 91.18 and then we've got the n which was 1,000 and that gives me the half width of the confidence interval and then adding and subtracting that gives us the upper and lower limits of the confidence interval. So it would say if this distribution is suitably modeled by a t distribution then we could be 95% confident that our profit would be between 154 and 143 but and that's a big but. Is this distribution something that's suitable to be modeled by the t or the normal distribution? Well the way to answer that question is to create a histogram and take a look. Now there's more complex methods you can use to actually see if a distribution is technically normal or technically t distribution but we're just going to look at the histograms then make a decision. And I'm going to zoom over here. I've built the histograms using the technique that we've seen and this profit distribution is obviously not normal, not a t distribution. So we'd be very careful using a standard distribution to model this and that's interesting if you look at the distribution of the profit versus the distribution of our random variable the demand if you remember we stay in this particular last model the demand was modeled using a uniform distribution and so this print out there looks fairly uniform so that tells us that the Monte Carlo simulation using that part but it just goes to show that your output variable may or may not parallel the distribution of one of your input variables and in particular if you've got multiple random variables say a uniform and a discrete and a normal and a triangular you've got all those input random variable distributions you really have to look at your output variable that you're concerned with to understand how to interpret the results since this is not a typical distribution we're going to have to just use the empirical way of coming up with some answers here and what I did I asked the questions what is the probability that the profit will be greater than two hundred and thirty dollars in this case and to find out we just use the count if function and in our range for profit here in our H column from H4 to H1003 that's our 1000 output values we run a test and if we're testing against a value we need to enclose it with the double quotation marks and here I wanted the test to see if whatever was in this profit sell is greater than two hundred and thirty dollars I wanted to count it and if it was not greater than two thirties to not count it and then to get the percent we divide that by 1000 and that came out to be thirty six point three percent or the count if brought back three hundred and sixty three places or outputs where the profit was greater than two thirty so we'd have a thirty six percent chance of making more than two hundred and thirty dollars if you wanted to find a range in this case someone said what it what is the probability of the profit being between two hundred and two hundred and thirty dollars and one way to do that is to get the use the count if again and what I did there I've got to count if statements I wanted to get the total count greater than two hundred dollars profit using the same method there and then subtract from that the count of profits greater than two thirty so that would get us the interval in between and of course just divide that by 1000 now I use the sum function here for that but you really wouldn't be do that if you're just using these two count ifs if you had multiple count ifs then the sum function would help you do that and then finally if you wanted to do a analysis of a column like I've added here where I've got a text output and I want to know the probability that I'm getting a negative profit you could do it of course with profit less than zero but here I just wanted to get a count of the number of times that negative showed up and we use our count if function again same well this time is the I column range where I put those text output and we count the number of times that negative shows up all you do is put in the double quotations again the value that the word that you want to return and you need to make sure that you're spelling it the same way here that you spelled it over here and I divided that by 1000 and got 7% chance of getting a loss with this particular model.