in the first calculation why did you take the square root of the number of days and not the square root of the product of 0,5*0,5/100 as you did exactly that in the next calculation?
25:23 question again sir, you said 50% chance failed, meaning 50% chance pass, but from the additional rule of probability, it should be 30%+10%-10%= 30% succeed or 70% failed. Is the 50% chance a predetermined number or am i missing something in calculation?
We are given three mutually exclusive events: that ONLY type 1 is successful, that BOTH type 1 and type 2 are successful or that ONLY type 2 is successful (any two of these outcomes cannot occur simultaneously), so I think he used the additional rule for mutually exclusive events.
Cash flow 400 for 15 years with discount rate 10%. DCF result is 3042,43 ; because the research itself already took 7 years, the DCF: 3042,43/(1+10%)^7 = 1561.24 . Now for the cost of research and FDA approval without drug production: 50 + 100/(1+10%) + 300/(1+10%)³ = 366.3 ; with production it cost: 366.3 + 600/(1+10%)^7 = 674.198. So the expected value is 1561.24 - 674.198 = 887.049
@@friednoodlee6599 Many thanks for your kind elaboration, can you advise for the 100M cost of testing for 2 years, why the formula is not to ^2. i.e why its not 100/(1+10%)^2 since the three year testing is 300/(1+10%)^3 ?
Thanks as always for the excellent teaching. Re the probability of 3 up days in a row (slide 5), why not 0.5569 x 0.5243 x 0.5243, taking into account the conditional probabilities from the data on slide 4, or am I missing something here?
I think since slide 4 shows that if the last day was up, it doesn't tell us a lot about the next day, we should just use regular probabilities and not conditional because the events seem independent. However, when the last day was down, the next day does become a dependent event and in that case we should use conditional probabilities. I'm not sure is this is right, but it makes sense..
Decision trees are a great tool to evade the sunk cost fallacy. Thanks, Professor.
Thank you so much for that decision tree explanation, Professor. You're the best!
24:40 ...what happens if it fails....the game's over!😂
in the first calculation why did you take the square root of the number of days and not the square root of the product of 0,5*0,5/100 as you did exactly that in the next calculation?
I think this was just a mistake like professor is calculating 500 days in mind.
Because all next calculations are according to formula.
So is the post class test, I didn’t get it either.
He made an error. Somehow he did 0.25/10. Which means he did 0.5sqrd/sqrt100
He forgot to take the square root of the numerator.
The standard error should’ve been 5% not 2.5%
20:15 i am surprised by the fact that there's no debt ratio or something similar in the formula. Is there any explanation in the paper?
Thank You Professor
Thanks!!!
25:23 question again sir, you said 50% chance failed, meaning 50% chance pass, but from the additional rule of probability, it should be 30%+10%-10%= 30% succeed or 70% failed. Is the 50% chance a predetermined number or am i missing something in calculation?
We are given three mutually exclusive events: that ONLY type 1 is successful, that BOTH type 1 and type 2 are successful or that ONLY type 2 is successful (any two of these outcomes cannot occur simultaneously), so I think he used the additional rule for mutually exclusive events.
A bit confused by the brexit slide 😅 esp where the number comes from … on their financial statement?
Can someone please explain to me how did the professor get the expected values in the decision tree ($887.05,-$366.3 etc)?
He used present value formula. He discounted cashflows, though I don’t what discount rate he used.
Cash flow 400 for 15 years with discount rate 10%. DCF result is 3042,43 ; because the research itself already took 7 years, the DCF: 3042,43/(1+10%)^7 = 1561.24 . Now for the cost of research and FDA approval without drug production: 50 + 100/(1+10%) + 300/(1+10%)³ = 366.3 ; with production it cost: 366.3 + 600/(1+10%)^7 = 674.198. So the expected value is 1561.24 - 674.198 = 887.049
@@friednoodlee6599 Many thanks for your kind elaboration, can you advise for the 100M cost of testing for 2 years, why the formula is not to ^2. i.e why its not 100/(1+10%)^2 since the three year testing is 300/(1+10%)^3 ?
@@friednoodlee6599 Thanks a lot :-)
anyone know how to get the cost of development and annual cash flow?
Thanks as always for the excellent teaching. Re the probability of 3 up days in a row (slide 5), why not 0.5569 x 0.5243 x 0.5243, taking into account the conditional probabilities from the data on slide 4, or am I missing something here?
0.5243 is for conditional probability, but in cumulative probability the events has to be independent. refer session 6 slide no. 4.
I think since slide 4 shows that if the last day was up, it doesn't tell us a lot about the next day, we should just use regular probabilities and not conditional because the events seem independent. However, when the last day was down, the next day does become a dependent event and in that case we should use conditional probabilities. I'm not sure is this is right, but it makes sense..
slide#5, third bullet has copy-paste bug.