Home     Forum     401k     401k Rollovers
    Register   Login   Members   Search   FAQs     Recent Posts    




Investment Comparison Spreadsheet

Reply to topic
Money Talk > Retirement Planning

Author Thread
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

Wino, I an not sure how much I can help you. you seem to think you know more than you really do and do not seem open to looking at what I am saying. that is fine by me, but I cannot communicate these points to you if you just keep sticking with your old ideas which bear no relationship with what I am saying.

for example, you are relating market LEVELS to changes in market levels needed for the TIPs analysis. your spreadsheet just compares apples to oranges. real estate may not be as directly related to inflation as you think. that said, it is a powerful hedge against your future cost of living, to the extent it is the same property you attempt to live in. also, financing it through fixed rate debt and not using oldguy's speculative strategies are also good ideas.

but I don't think I can tell you anything more...

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Mon Aug 19, 2013 2:18 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

quote:
Originally posted by smk
progress at last...

I guess your typing speed was finally s l o w enough. Laughing

~Tim~
Post Mon Aug 19, 2013 2:22 pm
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

quote:
Originally posted by coaster
I guess your typing speed was finally s l o w enough. Laughing


for you I think. what did you think of the next steps...as in any analysis, you start out developing simple approximations and continue to work towards making them realistic...

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Mon Aug 19, 2013 2:24 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

quote:
Originally posted by smk
but I don't think I can tell you anything more...

Frankly, I think the whole "failure to communicate" here is that Don and I have been looking at the macro, and Steve has been looking at the micro. Which is, itself, "apples and oranges" Wink

~Tim~
Post Mon Aug 19, 2013 2:24 pm
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

sorry, but you have to explain to me how it is I am not looking at the macro? perhaps this is part of the communication gap Wink

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Mon Aug 19, 2013 2:29 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

I have a story (personal experience) that illustrates my view on that but don't have the time to post it right now. If I haven't done so in a couple days, please remind me.

~Tim~
Post Tue Aug 20, 2013 5:11 am
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

well, wino is building a model to evaluate the macro situation. apparently his model seems to focus on maximizing nominal dollars at a particular point in time. I am building a model to evaluate the macro situation. my model includes both investment and non-investment factors and targets the objective of being able to actually pay your bills in retirement. the communication gap comes from an apparent disinterest in whether you can pay your bills in retirement in favor of the not so relevant ability to wave the flag to show how much money you have on a particular day...

sorry, but what I am doing is definitely macro.

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Tue Aug 20, 2013 10:43 am
 View user's profile Send private message
Wino
Senior Member


Cash: $ 113.80

Posts: 560
Joined: 03 Aug 2012
Location: Dubai
 Reply with quote  

quote:
Originally posted by smk
wino is building a model to evaluate the macro situation. apparently his model seems to focus on maximizing nominal dollars at a particular point in time.

emphasis added

A few points: Wino has built a model. I can change it, but the model is there, and it functions.

The model doesn't maximize or minimize anything. The model allows you to enter variables for two different future value equations for the purpose of comparing the outcome of the two strategies. Some of the variables are:

Amount invested initially (balloon, $)
Amount invested monthly ($)
Amount of any front end load fees (%)
Amount of any back end load fees (%)
Annual return (%)
Percentage for admin fees (%)
Annual admin fees ($)
Share purchase transaction fee ($)
Share sale fee ($)

Some fees or variables can be adjusted each year, but some cannot. If a change is made, that change is used from that point forward, while previous calculations use the previous variable.

So, where does this try to maximize anything? It doesn't. It allows you to compare two different investments assuming the same dollar amount is invested periodically in each of them. What variables you use are up to you, and you can use 0 or zero per cent for variables that do not apply to a given investment.

You want another variable? Name it, and define it. If you want to discuss nebulous trivia about government bonds, fine... define its variables, but I bet the figure is a percentage, which is already allowed in my model, and therefore can be used to model the results.[/quote]
Post Tue Aug 20, 2013 1:40 pm
 View user's profile Send private message
Wino
Senior Member


Cash: $ 113.80

Posts: 560
Joined: 03 Aug 2012
Location: Dubai
 Reply with quote  

quote:
Wino, I an not sure how much I can help you. you seem to think you know more than you really do and do not seem open to looking at what I am saying.


Let me help you by explaining how to use this model:

You invest money starting whenever you like. That is "year 1."

Do you want to use "real" variables? Then start with assumptions in 1992, and use variables from then until 2012 for any different number of scenarios.

Do you want to actually TRACK what you do today, compared to what you COULD HAVE done? Then start with assumptions in 2013, and update the variables annually. Was it a good choice you made?

Want to figure out which was good for the last ten years, then start investing now? Start with some figures in 2003, and figure out which one did better historically, and start with a "negative fee" in 2013. That's just like a deposit. You can then zero your previous investment variables and go forward from there.

Want to do two completely different investments? Fine. Change the percentages annually; change the amounts annually using the "fees" dollar which allows you to buy (negative fee) or sell (positive fee).

The good thing about custom models is I can make them do anything I want. I'll ask one more time: Do you want a different variable? Should you be able to change it annually? Should it be a dollar amount or a percentage of something; if a percentage, a percentage of what?

You seem to think that random variables I put in are some analysis I've done. I just commented on some of the results using some numbers pulled from a single prospectus against the SP500 on Vanguard, and then I said I was somewhat surprised by the results.

It sure is good to know what the results are when you're making decisions, and if you can vary some of your assumptions, then you can get a clearer picture, still. Good thing I have a model that allows me to do it, now that I've written it.
Post Tue Aug 20, 2013 2:02 pm
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

wino...I stopped doing stuff like your model (which I did in my sleep) in the early 1980's because it was too antiquated.

you need to enter a lot more variables. what about tax structure for roth, trad ira and maximizing the benefits based upon changing tax rates throughout your life. what about deferring pension income (social security based upon an assumed reduction in the amount or private pensions), what about the impact of longevity through lifetime annuities, what about inflation spikes, what about the costs of home ownership/downsizing and moving to a no tax state. these are all basic ideas on how to improve retirement income without taking market risk. if you don't consider these first, you are doing dw a disservice.

but before you get to these, you need to know how to evaluate whether your alternative investments are good or bad. your model output shows total money at the end of the period with both alternatives, but you can't decide which is better based upon just that. the return assumptions you use are not guaranteed. what about dispersion of returns. then, what are you comparing it to. for you it is nominal dollars at the end of the period. what about inflation adjusted dollars at the point in time when you actually need each dollar. then have you considered you can purchase that stream of cash flows in the open market and monitor the price every day? how does the price of your alternative investments match up to the price of the cash flows you will need when you actually retire on a daily basis? take the changes in returns and compare them to each other - what is the standard deviation (dispersion) of those returns. what is the dispersion of actual dollars you will have to spend every year in your retirement?

again, this is all fairly basic stuff. but you can't compare one asset to another without knowing how to evaluate which is better. your model produces dollars at a point in time as the output, but that point in time is not the point in time when you need it. also it does not consider the inflation of those dollars needed or the dispersion of returns verses those inflated dollars...

you wanted some variables...you got em Wink

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Tue Aug 20, 2013 2:36 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

Steve, I decided to not relate my story. It's too long. And in any case, we're just going to rehash ground you and I have covered before about the limits of models. I didn't say you weren't covering the macro. I did say you were looking at the micro. You want a model with a lot of micro inputs. And therein lies the problem. Any one of the micro input assumptions that in real life deviates from the assumption changes the result of the macro. And the more the inputs, the more they can gang up on skewing the resulting macro. The approach I've decided to take, after my own bout with models, is to postulate the macro based on logic, common sense, and real-life observations. Then, while implementing that macro, to observe and test the outcomes on an ongoing basis to determine if the macro holds. If the macro holds over time, it means my postulate was correct. If the macro fails over time, then my postulate was wrong, and I need to ditch it and develop another. It's more like real life. Things change and you adapt to the change. The models with all their micro assumptions give a sense of exactitude that simply isn't going to be there in real life.

The macro approach Don plans to take is real estate; this based on the logic of growing population demands for housing, the real-life observation that real estate appreciates over time, and the common sense that the income generated can cover the expense of ownership. It's a decent macro, proved over time, and you don't need any micro inputs or models to prove or disprove. You do need a generous margin of error in the macro to cover events like the recent real estate crash, but that's part of the logic and common sense, and doesn't need any models with multitudes of variables to figure that out.

~Tim~
Post Tue Aug 20, 2013 3:27 pm
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

quote:
Originally posted by coaster
Steve, I decided to not relate my story. It's too long. And in any case, we're just going to rehash ground you and I have covered before about the limits of models. I didn't say you weren't covering the macro. I did say you were looking at the micro. You want a model with a lot of micro inputs. And therein lies the problem. Any one of the micro input assumptions that in real life deviates from the assumption changes the result of the macro. And the more the inputs, the more they can gang up on skewing the resulting macro. The approach I've decided to take, after my own bout with models, is to postulate the macro based on logic, common sense, and real-life observations. Then, while implementing that macro, to observe and test the outcomes on an ongoing basis to determine if the macro holds. If the macro holds over time, it means my postulate was correct. If the macro fails over time, then my postulate was wrong, and I need to ditch it and develop another. It's more like real life. Things change and you adapt to the change.


we agree on the limits of models. I do the same was monitoring model performance against reality. the question is when you start with a model, are you targeting your objective or something else?

as for your approach on high dividend paying stocks, if you look at very long history, dividends produce the majority of economic benefit. however, in periods of high inflation, the underlying businesses cannot produce dividend improvements that match inflationary increases. common sense says your approach has a hole in it. an emphasis on reits as dividend payors may help, but again the deviations can be substantial. if you take out a long term fixed rate mortgage where the notional amount matches the amount of d paying stocks, the inflationary risk may be covered. if you are buying any bonds with limited credit risk and have not considered inflation protected assets instead, that is a very inefficient way to play the fixed income markets given a retirement objective. these are the common sense results that come from studying the models I was discussing above. but in order to get to the common sense, you need to start with a model that targets your objective and study how things work...wino is definitely not doing that...

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Tue Aug 20, 2013 3:39 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

"periods of high inflation", ya, but how long do those last? You have to look at the average over long periods of time. There will be short-terms periods of leading and short-term periods of lagging. Such is the nature of the beast. Over the long term it leads.

~Tim~
Post Tue Aug 20, 2013 3:41 pm
 View user's profile Send private message
smk
Preferred Member


Cash: $ 37.05

Posts: 183
Joined: 12 Dec 2012

 Reply with quote  

quote:
Originally posted by coaster
"periods of high inflation", ya, but how long do those last? You have to look at the average over long periods of time. There will be short-terms periods of leading and short-term periods of lagging. Such is the nature of the beast. Over the long term it leads.


this is a common sense question, not a modeling one. sometimes I use models to explain it though. you are assuming periods of high inflation by definition must be short. is there some fundamental reason to guaranty that or does that just happen to be the recent history in the US. Shocked

there are economies that experience hyperinflation. you can say it is short lived, but everything is wiped out by it. you might also find other countries with persistent inflation above 5%.

in your scenario if something like this happens you can get wiped out. what is the economic benefit for the risk? do you know?

for stories, I remember when people needed to come up with an index portfolio to match a mbs index, they would look at the variance-covariance matrix and simply optimize for the portfolio that minimized the tracking error. I never did that. I used a stratified sampling method when you took samples from all the coupon, maturity, and product ranges. take a few mbs from each box. there were reasons for their different behaviors even if it did not show up in the model, so you need to make sure you target the reasons for the behaviors. your approach here is like the first. mine is the second.

Steve Kanney, CFA
http://www.integratedfinancialny.com/index.html
Any comments made are designed to help you make your own decisions and do not consititute investment advice.
Post Tue Aug 20, 2013 4:03 pm
 View user's profile Send private message
coaster
Senior Advisor


Cash: $ 1626.30

Posts: 7990
Joined: 11 Oct 2005
Location: Wisconsin
 Reply with quote  

Ummmm, well, I don't have the academic background you do, and my academic mathematics was of the engineering variety, not the finance variety; so I don't have a command of the proper language you do, but based on your two approaches described above, I'd say my approach *when I do make a model* (which hasn't been for years) is probably actually closer to the second.

What I learned about models, I learned trading futures, because I traded a system, which is a model, and made pretty decent money off of it for a while, was that every model blows up sooner or later, and the more variables the model has, the sooner it's going to blow up. And that's from these people: http://www.futurestruth.com/ who are the futures community's system watchdogs. And the sytem I used did eventually fail (though by that time I had already moved on to something else) and it only had two inputs and only one of those was a variable.

~Tim~
Post Wed Aug 21, 2013 4:13 am
 View user's profile Send private message

Goto page Previous  1, 2, 3, 4  Next
Reply to topic
Forum Jump:
Jump to:  
  Display posts from previous:      





Money Talk © 2003-2016