While planning for retirement, we encounter several rates:
- rate of return before and after retirement
- rate of inflation before and after retirement
While calculating the corpus required and the monthly investment required, we assume these rates are constant for each year before retirement and during retirement.
In reality, all these rates can vary quite a bit. This means, the actual corpus required and therefore, the monthly investment required can also vary.
A Monte Carlo retirement calculator is one in which the above mentioned rates are varied randomly (but in accordance with a statistical distribution) several times, and the probability of financial independence of a given number of years is calculated.
In the previous version of the Monte Carlo calculator, I had presented answers in terms of the monthly investment instead of a probability. This was confusing and as pointed out by Paresh Girdhar, gave some confusing and erroneous results.
Therefore, I have replaced the confusing section with the probability of financial independence for a given number of years in retirement. Hopefully, the sheet is now simpler to use.
The probability of financial independence is determined by randomly varying the above-mentioned rates, 50,000 times in each trial as shown below (click for full size). The user can either use a normal distribution or an uniform distribution. If you are not comfortable about this, simply use the normal distribution for all quantities.
More details on Monte Carlo Simulations (disadvantages, alternatives etc.), can be found here and here
Download the updated Monte Carlo Simulator
Thanks to Rajesh for pointing out a bug.
Register for the
Chennai DIY Investor Workshop Jan 29th, 2017
Install Financial Freedom App! (Google Play Store)
Install Freefincal Retirement Planner App! (Google Play Store)
Buy our New Book!You Can Be Rich With Goal-based Investing A book by P V Subramanyam (subramoney.com) & M Pattabiraman. Hard bound. Price: Rs. 399/- and Kindle Rs. 349/-. Read more about the book and pre-order now! |
Fantastic work Pattabiraman. I am really impressed at the effort that you have put in to build all these different calculators. They are definitely quite relevant for practical personal financial planning. Considering that most calculators in the personal finance space in india are more of a theoretical and non-practical nature, your blog is indeed a breath of fresh air for people like me.
I am particularly impressed with this calculator for stress testing of retirement plan, considering very limited historical data that we can rely on here and the volatility we find in the rate of returns as well as inflation. Thank you for all this.
One small observation on the stress testing calculator is that i noticed the probability of success seems to be decreasing as you reduce the number of years in retirement which for me was counter intuitive. For e.g, i found that the probability of success for certain set of values of interest rates, inflation, distribution assumptions etc, for 45 years in retirement was 89%, however when i changed this value to 20 with all the other values remaining the same the probably of success reduced to 58%. Do let me know if i am missing something here.
Regards
Rakesh
Thank you. The issue is due to a bug. I have corrected it and uploaded a revised version with a credit to you. Many thanks.
Fantastic work Pattabiraman. I am really impressed at the effort that you have put in to build all these different calculators. They are definitely quite relevant for practical personal financial planning. Considering that most calculators in the personal finance space in india are more of a theoretical and non-practical nature, your blog is indeed a breath of fresh air for people like me.
I am particularly impressed with this calculator for stress testing of retirement plan, considering very limited historical data that we can rely on here and the volatility we find in the rate of returns as well as inflation. Thank you for all this.
One small observation on the stress testing calculator is that i noticed the probability of success seems to be decreasing as you reduce the number of years in retirement which for me was counter intuitive. For e.g, i found that the probability of success for certain set of values of interest rates, inflation, distribution assumptions etc, for 45 years in retirement was 89%, however when i changed this value to 20 with all the other values remaining the same the probably of success reduced to 58%. Do let me know if i am missing something here.
Regards
Rakesh
Thank you. The issue is due to a bug. I have corrected it and uploaded a revised version with a credit to you. Many thanks.
Wow, Great one. But I have a question on this. I just played around this with some numbers. For example for 40 years in retirement with Rs 55k per month in expenses, Rs 3.5cr gets 80% of probability. But I couldn't get 100% even if I increase to Rs 10 cr. I kept all other parameters for variations as it is. Not sure if I am doing anything wrong.
BTW at what is the practical % one should be comfortable with? Does one should be comfortable with 80% probability or 100% or less? I know more is merrier, but it needs more corpus or less years in retirement or roi should be higher or inflation should be lower. Can you let me know? Thanks
Thanks. I dont think 100% is possible because of the fluctuations. I think 80% is pretty decent.
Wow, Great one. But I have a question on this. I just played around this with some numbers. For example for 40 years in retirement with Rs 55k per month in expenses, Rs 3.5cr gets 80% of probability. But I couldn't get 100% even if I increase to Rs 10 cr. I kept all other parameters for variations as it is. Not sure if I am doing anything wrong.
BTW at what is the practical % one should be comfortable with? Does one should be comfortable with 80% probability or 100% or less? I know more is merrier, but it needs more corpus or less years in retirement or roi should be higher or inflation should be lower. Can you let me know? Thanks
Thanks. I dont think 100% is possible because of the fluctuations. I think 80% is pretty decent.