Dear Reader:
Welcome to Quantitative Singularity, a site devoted to explore modelling of practical financial planning scenarios. As the name implies, it relies on using quantitative methods that range from the simple to more complex as we examine various strategies and guidelines. While the focus is primarily on retirement issues themes explored are of sufficiently general nature. Some individuals most likely have engaged the services of financial advisors to help them in these activities. It is not intention here to replicate any of those. Rather it is to provide knowledge, information and tools to independently assess one’s situation and perhaps even to design and execute strategies based on repeatable procedures.
The software used is the R language, a freely available tool widely used for data analysis and data mining. It is often enhanced by the addition of external packages which are also practically all open source and freely available.
Various articles, papers and posts will be posted that are of interest. In addition, implementing strategies or models will be suggested through code in R. Many snippets will be supplied and built over time.
Plotting a safe path through retirement has been the object of serious study and analysis especially given volatility of markets, uncertainty of events, unpredictability of future expenses and income streams. These are not the only variables that enter into the equation. Research conducted across many different overlapping disciplines has resulted in a vast available literature and increasingly esoteric. So, where does one begin.
At the beginning of course, as the Mad Hatter said to Alice. Conducting computations across large data sets and many simulations of possible scenarios, extracting and presenting meaningful information from the results in a timely manner, all point to using applicable software toolsets and toolkits. Software and algorithm design must of necessity be based on proven statistical and mathematical techniques as applied to economics and finance. So, clearly that indicates at least 2 major areas requiring attention that become necessary and hopefully sufficient for our purposes. It should be noted that analysis should be repeatable and demonstrable, based on empirical data, which immediately eliminates all ‘from the gut’ like approaches.
Above all, these must be simple to understand and implement.
Software and associated toolkits are the first area of requisite expertise and learning to program or code in any language is a steep climb at the best of times. However, this is not as dissuading as one might originally suppose. Maintaining / solving home budgeting problems in Excel is undertaken universally and one has in a sense already begun the climb. R takes it to a whole other level but the principle is the same nevertheless.
The other area comprises the statistical and mathematical techniques that are necessary to model or simulate prices, account for uncertainty of events, determine optimal allocations, analyze sensitivity of variables to desired outcomes or the avoidance of disaster and so on. I am glad to note that there is indeed an excellent book, available freely, referenced on this site that fortunately seems to tie all this together. It is called “Financial Analytics with R – Building a Laptop Laboratory for Data Science”. It starts off with a brisk introduction to R and then fairly gallops through several critical concepts with a sprinkling of the underlying math or statistics. The really cool part is that it abounds with sample R code that one can experiment with on the way.
It would appear at first glance that some material referenced on this site is just too much math or Greek symbols. These may be conveniently skipped for those time-challenged to follow each derivation or proof. It is important though to understand the proposition or problem that is always clearly presented and discussed first in plain language. The resort to excessive math underlines the logic of the approach, and only serves to enhance one’s confidence in the techniques suggested or used. Often R code is helpfully supplied with clear enough explanations. That should be sufficient for our uses and in fact immensely helpful in plotting our own financial paths forward.
Mention should be made of the data required to undertake any type of analysis. This is a critical area and helpful hints are provided as to where free data can be obtained. The Feds of course maintain enormous amounts of market data available to the public.
One final comment. I invite readers to provide feedback, suggestions or corrections to make this site better. The aim is to provide peer-reviewed information that can be confidently used in planning real-life situations with real consequences. Please use comments in blog, i.e. Posts/Discussions area for any responses.
That’s it for now. Lets get it off to a start. It would be helpful if both the R-language and an Integrated Development Environment (IDE) such as R-Studio is downloaded and installed.
Links are provided in the appropriate category.
Happy cruising on the quantit-analytic highway!