Hello all,
I just received my DNMMR on my first attempt at the FA and wanted to get a sense of what common mistakes I might be making. I've listed below my initial thoughts on why I DNMMR, and would love any feedback, especially from those that did MMR.
Overall
1. Being too wordy - mine came in around 10k words without tables, so I'm guessing I need to shorten it up and be more concise
2. Be more black and white in my recommendation. I probably put too many caveats and limitations on how applicable my price range was.
3. Went beyond the scope of questions in trying to consider all the risks of each task
Task 1 Specifics
1. Focused too much on frequency of errors as opposed to severity
2. Probably was unreasonable in considering how much it would actually take to fix any data errors
Task 2 Specifics
1. Maybe my range was too aggressive? I had a walkaway point using a risk metric that was at -10%?
2. I thought I did a solid job justifying my performance and risk metric (e.g. why PVP/IR instead of just PVP, why PVP/IR >10% instead of PVP/IR >0%, why CTE instead of VaR or StdDev, why CTE(90) instead of CTE(75) or CTE(95), why CTE(90) > -10% instead of CTE(90) > 0%), but maybe I was too detailed?
3. Maybe too detailed in my risk/reward analysis of pricing. I could have just said higher price = higher profit expectation, but also less likelihood of winning the work
4. Maybe should have just said a high and low price point instead of outlining 5 specific points along that range?
Task 3 Specifics
1. I only tested each assumption at my walkaway point, maybe should have tested it at the high and low ends?
2. I only tested each assumption with an X% and 2X% shock in either direction - maybe it makes more sense to have different shocks for different assumptions?
3. Maybe mention more about interdependence of assumptions? Maybe discuss anti-selection?
4. Have better factors excluded that I think are material? I came up with 3 but maybe that's not enough?
Task 4 Specifics
1. I didn't attempt to quantify the severity or likelihood of my key risks, but mentioned ways that we could do that. Maybe I should try to estimate those?
2. I listed more operational than strategic or financial risks in my RCD - maybe I need a more balanced count?
3. Maybe my key risks were too broad? For example, "Decreased credit usage leading to fewer miles purchased" could instead have been "Increase in penalty rates on CDL cardholders leading to fewer miles earned and a decrease in volume of Air Miles purchased"?
Task 5 Specifics
1. Maybe focusing too much on risk/reward of each policy change and not the risk/reward of implementing it in the model?
2. Should I have focused more on changes to the model and less on why each policy itself should have been changed or not?
Task 6 Specifics
1. Probably too wordy
2. Probably not clear enough of a recommendation (too many caveats on data quality, understanding of current risk appetites, etc.).
3. I'm guessing I should focus more on presenting a strong argument for why my range should be justified and leave out most of the limitations or discussions of limitations on my price range? I don't know what the appropriate balance is between providing a black and white answer and providing the most appropriate and accurate response that adequately discusses the risk associated with this price range?
4. Maybe consolidate my tables more into the report instead of just having them in an appendix at the end?
Task 7 Specifics
1. I made specific recommendations for assumption changes (e.g. Decrease the miles purchased assumption to 12 million miles). Should I have been less specific, and just said "Decrease the miles purchased assumption"?
2. I made mention of updating the model to match final deal structure, maybe this was beyond the scope of the question?
Like I said, any advice or discussion would be helpful (since we don't get any feedback until after the 2nd failed attempt). I'm fairly frustrated with myself, and the SOA for providing no feedback or model solution or anything. Thanks in advance, congratulations to all those who have MMR, and for those who DNMMR - I'll be retaking this on 2/19-2/23 if you want to join!
I just received my DNMMR on my first attempt at the FA and wanted to get a sense of what common mistakes I might be making. I've listed below my initial thoughts on why I DNMMR, and would love any feedback, especially from those that did MMR.
Overall
1. Being too wordy - mine came in around 10k words without tables, so I'm guessing I need to shorten it up and be more concise
2. Be more black and white in my recommendation. I probably put too many caveats and limitations on how applicable my price range was.
3. Went beyond the scope of questions in trying to consider all the risks of each task
Task 1 Specifics
1. Focused too much on frequency of errors as opposed to severity
2. Probably was unreasonable in considering how much it would actually take to fix any data errors
Task 2 Specifics
1. Maybe my range was too aggressive? I had a walkaway point using a risk metric that was at -10%?
2. I thought I did a solid job justifying my performance and risk metric (e.g. why PVP/IR instead of just PVP, why PVP/IR >10% instead of PVP/IR >0%, why CTE instead of VaR or StdDev, why CTE(90) instead of CTE(75) or CTE(95), why CTE(90) > -10% instead of CTE(90) > 0%), but maybe I was too detailed?
3. Maybe too detailed in my risk/reward analysis of pricing. I could have just said higher price = higher profit expectation, but also less likelihood of winning the work
4. Maybe should have just said a high and low price point instead of outlining 5 specific points along that range?
Task 3 Specifics
1. I only tested each assumption at my walkaway point, maybe should have tested it at the high and low ends?
2. I only tested each assumption with an X% and 2X% shock in either direction - maybe it makes more sense to have different shocks for different assumptions?
3. Maybe mention more about interdependence of assumptions? Maybe discuss anti-selection?
4. Have better factors excluded that I think are material? I came up with 3 but maybe that's not enough?
Task 4 Specifics
1. I didn't attempt to quantify the severity or likelihood of my key risks, but mentioned ways that we could do that. Maybe I should try to estimate those?
2. I listed more operational than strategic or financial risks in my RCD - maybe I need a more balanced count?
3. Maybe my key risks were too broad? For example, "Decreased credit usage leading to fewer miles purchased" could instead have been "Increase in penalty rates on CDL cardholders leading to fewer miles earned and a decrease in volume of Air Miles purchased"?
Task 5 Specifics
1. Maybe focusing too much on risk/reward of each policy change and not the risk/reward of implementing it in the model?
2. Should I have focused more on changes to the model and less on why each policy itself should have been changed or not?
Task 6 Specifics
1. Probably too wordy
2. Probably not clear enough of a recommendation (too many caveats on data quality, understanding of current risk appetites, etc.).
3. I'm guessing I should focus more on presenting a strong argument for why my range should be justified and leave out most of the limitations or discussions of limitations on my price range? I don't know what the appropriate balance is between providing a black and white answer and providing the most appropriate and accurate response that adequately discusses the risk associated with this price range?
4. Maybe consolidate my tables more into the report instead of just having them in an appendix at the end?
Task 7 Specifics
1. I made specific recommendations for assumption changes (e.g. Decrease the miles purchased assumption to 12 million miles). Should I have been less specific, and just said "Decrease the miles purchased assumption"?
2. I made mention of updating the model to match final deal structure, maybe this was beyond the scope of the question?
Like I said, any advice or discussion would be helpful (since we don't get any feedback until after the 2nd failed attempt). I'm fairly frustrated with myself, and the SOA for providing no feedback or model solution or anything. Thanks in advance, congratulations to all those who have MMR, and for those who DNMMR - I'll be retaking this on 2/19-2/23 if you want to join!
FA DNMMR Common Mistakes