Predicting future Net Worth
 steelerfan
 Posts: 169
 Joined: Fri Mar 13, 2009 10:51 am
 Location: near Savannah, GA
Predicting future Net Worth
So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
"You make most of your money in a bear market, you just don't realize it at the time."  Shelby Cullom Davis
Re: Predicting future Net Worth
It sounds like your trying to plot 8 data points and use regression analysis to plot a trend line?
I think you'd be better off to project how much you will save, how much debt you can/will pay off, and use a a range of projections for returns on your assets.
5 years is awfully short for projecting asset returns. The range will be (and should be) really large. For fun, maybe look at 5% to 15% for your portfolio. Just know that its just a guess and for entertainment purposes only.
I think you'd be better off to project how much you will save, how much debt you can/will pay off, and use a a range of projections for returns on your assets.
5 years is awfully short for projecting asset returns. The range will be (and should be) really large. For fun, maybe look at 5% to 15% for your portfolio. Just know that its just a guess and for entertainment purposes only.
Re: Predicting future Net Worth
A caveat regarding net worth projections based on personal experience. Based on assets with Vanguard, we took advantage of a free financial plan based on extensive information regarding present expenses, income, assets and debt at the time, even projected times between automobile purchases, upcoming college expenses for children, etc. In other words a significant task to obtain the necessary information which we did and submitted accurately.
The neatly completed and bound report reflected a multitude of financial projections from the year 2000 through 2037 including total portfolio balances and inflation adjusted portfolio balances. Approximately 17 years later, we have followed that plan quite closely in terms of expenditures and investing. The current result is that our total portfolio and inflation adjusted portfolio balances are each almost exactly one half of the projections! That is quite a negative difference when you consider the apparent sophistication of the process undertaken and our adherence to the recommendations. Fortunately, the actual amounts are sufficient to allow a comfortable retirement even while wistfully looking at the projected amounts.
Tim
The neatly completed and bound report reflected a multitude of financial projections from the year 2000 through 2037 including total portfolio balances and inflation adjusted portfolio balances. Approximately 17 years later, we have followed that plan quite closely in terms of expenditures and investing. The current result is that our total portfolio and inflation adjusted portfolio balances are each almost exactly one half of the projections! That is quite a negative difference when you consider the apparent sophistication of the process undertaken and our adherence to the recommendations. Fortunately, the actual amounts are sufficient to allow a comfortable retirement even while wistfully looking at the projected amounts.
Tim

 Posts: 3009
 Joined: Tue Jul 26, 2011 1:35 pm
Re: Predicting future Net Worth
If you are near retirement, and the bulk of your networth is invested (not in a can in the backyard)
You should expect compounded returns, so exponential would be more correct than linear.
There are several calculators out there (firecalc for example) which may be better than
making your own.
You should expect compounded returns, so exponential would be more correct than linear.
There are several calculators out there (firecalc for example) which may be better than
making your own.
Re: Predicting future Net Worth
I split out each of my retirement accounts in Excel (separate columns): 401k, Taxable, IRA, Pension
I have a starting point at the top of each column (current value).
For each account, I have separate columns for annual contributions, company match and expected return (I currently use 5% annual).
I then add up each year (I forecast out to 67 years old) and roll forward the balance year over year.
Then I add up each account's forecast value to arrive an aggregate combined value year over year.
I don't add home equity.
I have a starting point at the top of each column (current value).
For each account, I have separate columns for annual contributions, company match and expected return (I currently use 5% annual).
I then add up each year (I forecast out to 67 years old) and roll forward the balance year over year.
Then I add up each account's forecast value to arrive an aggregate combined value year over year.
I don't add home equity.
steelerfan wrote:So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
Re: Predicting future Net Worth
You need to split net worth growth into market appreciation and savings. Market appreciation will apply exponentially (growth rate based on historical averages for asset allocation), savings are likely to be flat or linear.steelerfan wrote:So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
Exponential extrapolations will be extremely noisy and will overestimate your future wealth, and a linear extrapolation will underestimate your future wealth. You should view them as upper/lower bounds or take the average of the two.
Re: Predicting future Net Worth
Wow. Very impressive. Deviation from plan is most overwhelmingly influenced by unforeseeable variables. One could have modified one's original plan with alternate scenarios to generate an expectation, but then the uncertainty would reside in the probabilities assigned to the scenarios. Only in hindsight would the probabilities be known: 100% for what happened and 0% for all others. And one likely didn't imagine what really happened anyway.Nowizard wrote:A caveat regarding net worth projections based on personal experience. Based on assets with Vanguard, we took advantage of a free financial plan based on extensive information regarding present expenses, income, assets and debt at the time, even projected times between automobile purchases, upcoming college expenses for children, etc. In other words a significant task to obtain the necessary information which we did and submitted accurately.
The neatly completed and bound report reflected a multitude of financial projections from the year 2000 through 2037 including total portfolio balances and inflation adjusted portfolio balances. Approximately 17 years later, we have followed that plan quite closely in terms of expenditures and investing. The current result is that our total portfolio and inflation adjusted portfolio balances are each almost exactly one half of the projections! That is quite a negative difference when you consider the apparent sophistication of the process undertaken and our adherence to the recommendations. Fortunately, the actual amounts are sufficient to allow a comfortable retirement even while wistfully looking at the projected amounts.
Tim
I think the only answer is to redo one's plan periodically using the most conservative likely scenarios and hope for the best. I think mechanized spreadsheets only prove that we don't know what we don't know  and thus are better than nothing.
Re: Predicting future Net Worth
Looking back at some nice graphical projections of my net worth from the past show me with much less in reality than those rosy projections. Makes me wonder about the usefulness of the various retirement calculators.
 steelerfan
 Posts: 169
 Joined: Fri Mar 13, 2009 10:51 am
 Location: near Savannah, GA
Re: Predicting future Net Worth
OP here.
Thanks everyone for your input. So about 1/2 of my net worth is home equity and other property and the other half is in investments. So maybe the advice to split the difference between linear and exponential might be sound.
I plan to update my spreadsheet annually, so it will become more accurate as I get closer to my retirement. Plus it will become evident which projection is better.
Thanks everyone for your input. So about 1/2 of my net worth is home equity and other property and the other half is in investments. So maybe the advice to split the difference between linear and exponential might be sound.
I plan to update my spreadsheet annually, so it will become more accurate as I get closer to my retirement. Plus it will become evident which projection is better.
"You make most of your money in a bear market, you just don't realize it at the time."  Shelby Cullom Davis

 Posts: 2979
 Joined: Thu Aug 09, 2012 10:54 am
Re: Predicting future Net Worth
The retirement projections at my company say I'll have like 400k a year in retirement not including taxable accounts. I don't believe that number since I'm so far out. IMO these kind of projections aren't really valuable until your very near retirement and your expeses are known. I don't have kids yet and have no idea what a kid costs but I know I can afford it. That right there makes just about every calculator invalid for me.

 Posts: 91
 Joined: Tue Dec 27, 2016 12:59 pm
Re: Predicting future Net Worth
yepMathWizard wrote: There are several calculators out there (firecalc for example) which may be better than
making your own.
Re: Predicting future Net Worth
They are both wrong but the exponential one is more wrong.steelerfan wrote:So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
Between 1900 and 1999 (just to keep it an even 100 years), the S&P 500 grew at 11.8% on average. You can plug that into Excel and find that
Code: Select all
=power(1 + 11.8%, 100)
=73900.30458
Nope. It actually only turned into $16,417. You have to take volatility into account, otherwise you dramatically overstate your final portfolio value.
Re: Predicting future Net Worth
Agree that an exponential function is a very poor choice for making accurate financial predictions.
A better approach would be a Monte Carlo simulation method where a number of variables are randomly adjusted.
The result will be range of possibilities which are dependent on assumption and plans as well as chance.
Staying the course is of course the plan.
A factor of 2 (twice or half) over the course of 17 years is probably typical.
A better approach would be a Monte Carlo simulation method where a number of variables are randomly adjusted.
The result will be range of possibilities which are dependent on assumption and plans as well as chance.
Staying the course is of course the plan.
A factor of 2 (twice or half) over the course of 17 years is probably typical.
Re: Predicting future Net Wort
For net worth calculation purposes I carry my personal residences at the lesser of purchase price and fair market value. This is a conservative approach that recognizes it is difficult to monetize gains in personal residences. When I do think about the embedded games, I make sure to deduct my transfer taxes, capital gains and brokerage commissions. I also assume zero appreciation when projecting future net worth. I do all of this to maintain a conservative balance sheet.steelerfan wrote:OP here.
Thanks everyone for your input. So about 1/2 of my net worth is home equity and other property and the other half is in investments. So maybe the advice to split the difference between linear and exponential might be sound.
I plan to update my spreadsheet annually, so it will become more accurate as I get closer to my retirement. Plus it will become evident which projection is better.
For my financial assets, the number I find most interesting is the ratio of my projected annual savings to projected annual investment return on my portfolio. My projected annual savings is equal to gross income minus taxes minus living expenses. Note that I include in savings any money applied toward debt reduction because whether I apply savings to debt reduction or investment enures positively to my net worth.
My projected annual investment return on my portfolio is just an assumption based on my asset allocation. As John Bogle has taught us, the only things we control are taxes and expenses on our investment portfolio. I might assume 7% long term annual return, but as we all know that is a guess in a random walk down Wall Street.
I then combine my annual projected savings and annual projected investment returns and calculate how quickly my net worth will double based upon that accumulation rate.
Over the years I've had fluctuations both to the upside from unexpected income and to the downside in the form of challenging market returns, but in each rolling 3 to 5 year periods over the past 20+ years I have been fairly accurate in projecting increase in my net worth.
As one's annual savings decreases as a percentage of overall portfolio, net worth will fluctuate to a greater degree based upon asset allocation and investment returns. Once in retirement that will entirely be the case plus one needs to take into account spending down a portfolio as retirement income.
As I see it you can only control three things: savings rate, minimizing taxes and minimizing expenses
If your outgo is greater than your income, your upkeep will be your DOWNFALL.
Re: Predicting future Net Worth
This thread is now in the Personal Finance (Not Investing) forum (Net worth).
Re: Predicting future Net Worth
I'm not sure about the utility of including your personal residence in your net worth for retirement purposes unless you plan to either cash out and move towards a lower COL area or take out a reverse mortgage.
Re: Predicting future Net Worth
I agree with previous posters on firecalc for what you're trying to do.
I do run regression lines on my NW history also, it's interesting if not a particularly useful predictor. See what Excel shows as your rsquared values for the linear and exponential fit.
I do run regression lines on my NW history also, it's interesting if not a particularly useful predictor. See what Excel shows as your rsquared values for the linear and exponential fit.
Re: Predicting future Net Worth
Maybe I missed it, but I did not see Financial Engines mentioned. That may be helpful. I used to use it, but I am passed the time when it is really relevant, and I don't need it now.
My experience, dating from 1978, is that you can get long stretches (1020 years) when the return stays low, and long stretches when it remains high. For me, the span of years between 1980 and 2000 determined my retirement wealth. Basically, not much has happened in the last 15 years. My experience was a Dow Change from 750 to 14000. Contrast that to the change between the year 2000 and the present.
My opinion is that you should take Financial Engines seriously, and pay particular interest to the tails of the distribution, because if you think you cannot end up there, you would be wrong. I had one shot and I was very lucky.
My experience, dating from 1978, is that you can get long stretches (1020 years) when the return stays low, and long stretches when it remains high. For me, the span of years between 1980 and 2000 determined my retirement wealth. Basically, not much has happened in the last 15 years. My experience was a Dow Change from 750 to 14000. Contrast that to the change between the year 2000 and the present.
My opinion is that you should take Financial Engines seriously, and pay particular interest to the tails of the distribution, because if you think you cannot end up there, you would be wrong. I had one shot and I was very lucky.
Re: Predicting future Net Worth
If we have a 50% drop in 2017 and it takes 5 years to recover what does that do to your projection?
Re: Predicting future Net Worth
Is this a serious question or illustrating the point that NW projections are of very limited value?Nate79 wrote:If we have a 50% drop in 2017 and it takes 5 years to recover what does that do to your projection?
NW is inherently subject to change, up and down, based upon a number of factors.
A 50% drop in equities for 5 years in my case would be extremely beneficial to my NW in 2022. Why? I project to be a significant buyer during the next 5 years
If your outgo is greater than your income, your upkeep will be your DOWNFALL.
Re: Predicting future Net Worth
As always it comes down to why are you asking?
That will tell you how sophisticated you want to be, what you want to include, do you want an average answer, a what if assets do poorly answer, a worse case answer, do you want a simple point estimate or a probability distribution.
That will tell you how sophisticated you want to be, what you want to include, do you want an average answer, a what if assets do poorly answer, a worse case answer, do you want a simple point estimate or a probability distribution.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Re: Predicting future Net Worth
Mostly serious because the OP is projecting out about 10 years towards retirement. A downturn would have a serious effect on a NW projection if using some simple projection like this. Also, likely for the OP who is nearing retirement the amount they are contributing is small compared to the total NW so it will not have that much affect to buy low during the downturn. If the market is flat for 5 years or even is negative the OP's NW may be flat from now until retirement.NYCguy wrote:Is this a serious question or illustrating the point that NW projections are of very limited value?Nate79 wrote:If we have a 50% drop in 2017 and it takes 5 years to recover what does that do to your projection?
NW is inherently subject to change, up and down, based upon a number of factors.
A 50% drop in equities for 5 years in my case would be extremely beneficial to my NW in 2022. Why? I project to be a significant buyer during the next 5 years
 steelerfan
 Posts: 169
 Joined: Fri Mar 13, 2009 10:51 am
 Location: near Savannah, GA
Re: Predicting future Net Worth
OP here again.
Thanks everyone for chiming in. I did visit the firecalc website but did not see anything about Net Worth projections. It looks to me that it would be very helpful in answering the question, "will my retirement money last as long as I do?" Maybe someone could help me if I missed something there.
A little more information about me is probably needed.
My personal financial goal is to retire with a Net Worth > $1 million. I currently invest about 20% of my gross household income and am debt free except for my house (which will be paid off during the year I plan to retire). My current NW is $440k. My NW has almost tripled in the last 10 years and am hoping it will triple in the next 10 years! If it does, then my goal is achieved, if it only doubles, then of course I will not have achieved it.
So, there you have it!
Thanks everyone for chiming in. I did visit the firecalc website but did not see anything about Net Worth projections. It looks to me that it would be very helpful in answering the question, "will my retirement money last as long as I do?" Maybe someone could help me if I missed something there.
A little more information about me is probably needed.
My personal financial goal is to retire with a Net Worth > $1 million. I currently invest about 20% of my gross household income and am debt free except for my house (which will be paid off during the year I plan to retire). My current NW is $440k. My NW has almost tripled in the last 10 years and am hoping it will triple in the next 10 years! If it does, then my goal is achieved, if it only doubles, then of course I will not have achieved it.
So, there you have it!
"You make most of your money in a bear market, you just don't realize it at the time."  Shelby Cullom Davis
Re: Predicting future Net Worth
Could you retire with a net worth of $1M if your house was worth $1M and you had no other funds? This is the danger of including home value in your calculations.
Re: Predicting future Net Worth
Just take your portfolio now and multiple by (1 + rate)^10, add next years new money raised to the power (1 + rate)^9 years, do the same for the new money a year later times (1 + rate)^8, etc. Pick whatever rate you want, 2%, 4%, as you think you might get. Try a range to see what might happen (little mini Monte Carlo).
Easy to do in a spreadsheet. There are built in functions too, but you will learn more if you do it "long hand".
Project home value or whatever in a similar manner.
Add them up.
Easy to do in a spreadsheet. There are built in functions too, but you will learn more if you do it "long hand".
Project home value or whatever in a similar manner.
Add them up.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Re: Predicting future Net Worth
This! Or you could just take the average and call it a day. If less than 250k difference is going to seriously derail your plans, you need to save more or work an extra year or two.691175002 wrote:You need to split net worth growth into market appreciation and savings. Market appreciation will apply exponentially (growth rate based on historical averages for asset allocation), savings are likely to be flat or linear.steelerfan wrote:So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
Exponential extrapolations will be extremely noisy and will overestimate your future wealth, and a linear extrapolation will underestimate your future wealth. You should view them as upper/lower bounds or take the average of the two.
Does that include all your assets? What sources of income are you counting on in retirement? SS? Pension? Other? A 4% SWR on $1M is $40k. Will that income stream be sufficient for your retirement expenses? What is your current annual budget/expenses?steelerfan wrote:My personal financial goal is to retire with a Net Worth > $1 million. ... My current NW is $440k.
I would not count on more than a doubling in the next 8 years. We could get lucky and get a couple of 2030% years or avoid a minus 25% year, but combined, that would amount to less than 10% annualized. I think you need a contingency if you don't have enough to start retirement at your desired date.
Re: Predicting future Net Worth
It seems that linear and exponential curves are a pretty inappropriate way to predict future net worth. Net worth doesn't follow neat growth curves. You'd be better off examining your anticipated expenses and incoming money. Even then it's very hard to predict. It would basically require a crystal ball.steelerfan wrote:So, I have almost 8 years worth of net worth data, but would like the collectives' input on the best and most accurate way to predict future net worth growth. I have used Excel to do a Linear prediction and an Exponential prediction of each year from 2018 through 2026 (the year I want to retire). With the linear one I get about $500K less than the exponential one.
My questions are: Which method is better for this type of projection? Is there a better method to use for this? What do the Financial Advisors use?
Re: Predicting future Net Worth
The only graph it shows is "yearbyyear portfolio balances", which is exactly what a net worth projection is. If you're not retired yet, you can even go to the "Not Retired?" tab and put extra information there.steelerfan wrote:I did visit the firecalc website but did not see anything about Net Worth projections.
 steelerfan
 Posts: 169
 Joined: Fri Mar 13, 2009 10:51 am
 Location: near Savannah, GA
Re: Predicting future Net Worth
Yes  that includes all my assets. I plan to sell my home when I retire and pay cash for my retirement home.inbox788 wrote:Does that include all your assets? What sources of income are you counting on in retirement? SS? Pension? Other? A 4% SWR on $1M is $40k. Will that income stream be sufficient for your retirement expenses? What is your current annual budget/expenses?
So I am currently receiving a pension as a US Army Retiree and will also receive a Civil Service Retirement in addition to SS. I believe all of these do adjust for inflation as well. So I think my income stream will be sufficient in retirement.
"You make most of your money in a bear market, you just don't realize it at the time."  Shelby Cullom Davis
Re: Predicting future Net Worth
On that graph, note especially the wide range of final values, corresponding to starting the simulation in different years. The default initial values ($750,000 portfolio, spending $30,000 per year for 30 years, no other income or expenses) give final values ranging from about $300,000 (negative!) to about +$4,000,000, with some clustering in the ballpark of +$300,000 to +$500,000.AlohaJoe wrote:The only graph it shows is "yearbyyear portfolio balances", which is exactly what a net worth projection is. If you're not retired yet, you can even go to the "Not Retired?" tab and put extra information there.steelerfan wrote:I did visit the firecalc website but did not see anything about Net Worth projections.
My investing princiPLEs do not include absolutely preserving princiPAL.
Re: Predicting future Net Worth
Switching homes doesn't really change net worth, just shuffle amounts around.steelerfan wrote:Yes  that includes all my assets. I plan to sell my home when I retire and pay cash for my retirement home.inbox788 wrote:Does that include all your assets? What sources of income are you counting on in retirement? SS? Pension? Other? A 4% SWR on $1M is $40k. Will that income stream be sufficient for your retirement expenses? What is your current annual budget/expenses?
So I am currently receiving a pension as a US Army Retiree and will also receive a Civil Service Retirement in addition to SS. I believe all of these do adjust for inflation as well. So I think my income stream will be sufficient in retirement.
You're going to have 3 pensions! Excellent! Looks like any one of them could help you sustain a basic living in retirement. Having all 3 should be more than sufficient and a very nice retirement. Congratulations! You might not be able to spend it all, so maybe it's time to start looking at that fishing boat.
As discussed above, doing a piecewise model, separating the linear (income) from exponential (investment) elements is probably the most accurate method. But missing elements include home appreciation (exponential), pensions (zero NW, but growing in value from appreciation and contributions), and other incomplete accounting.
Did you pay off any debts (aside from mortgage) in the last 8 years? For a full accounting, debts might be considered negative bonds, and any retirement of debt is same as adding to bond net worth. If the cash stream from those debt payments was not completely counted in the past, but continues in the present, it provides good positive surprise.
In your case, if you project NW into retirement, I wouldn't be surprised if the exponential curve turns out to be more accurate.
 Aptenodytes
 Posts: 3751
 Joined: Tue Feb 08, 2011 8:39 pm
Re: Predicting future Net Worth
It helps to be as specific as possible in articulating the decisions you will be making that depend on your net worth estimate. Otherwise you won't formulate the indicator properly. You may discover that nothing depends on it and can save yourself some time.
There's an interesting book by Doug Hubbard, How to Measure Anything. He says the failings of most measurement systems is that they don't think carefully enough about how the metrics will be used in support of specific decisions, and they rely too much on guesswork to work out which metrics will do the best job. The implication is that all of us are probably using seriously suboptimal portfolio tracking systems.
The only reason to project the value of your house is if you think the cost of your new house will change in a very poorly correlated manner.with respect to your current house. If you think they are likely to move in similar ways you can just omit it from the calculus
I track net worth once a year to provide a motivational pat on the back when we are doing things right and to trigger sustained discipline if we slip.
I project my portfolio balances to support a range of decisions having to do with level of voluntary 403b contributions, choice of AA, whether to spend from HSA or allow it to accumulate. The future house price just doesn't matter for those things so I don't bother projecting it.
You could argue that I did a lightning projection of the difference in rate of growth in my home's value and real estate prices in likely locations for our postretirement home. The result was 'highly likely to permit buying almost all the house we want' so then I stopped pursuing any greater precision. I check the assumption's validity about once a year or so.
There's an interesting book by Doug Hubbard, How to Measure Anything. He says the failings of most measurement systems is that they don't think carefully enough about how the metrics will be used in support of specific decisions, and they rely too much on guesswork to work out which metrics will do the best job. The implication is that all of us are probably using seriously suboptimal portfolio tracking systems.
The only reason to project the value of your house is if you think the cost of your new house will change in a very poorly correlated manner.with respect to your current house. If you think they are likely to move in similar ways you can just omit it from the calculus
I track net worth once a year to provide a motivational pat on the back when we are doing things right and to trigger sustained discipline if we slip.
I project my portfolio balances to support a range of decisions having to do with level of voluntary 403b contributions, choice of AA, whether to spend from HSA or allow it to accumulate. The future house price just doesn't matter for those things so I don't bother projecting it.
You could argue that I did a lightning projection of the difference in rate of growth in my home's value and real estate prices in likely locations for our postretirement home. The result was 'highly likely to permit buying almost all the house we want' so then I stopped pursuing any greater precision. I check the assumption's validity about once a year or so.