Fair warning, I’m getting into topics I know less about. While I was once up to speed on modeling, I’m really getting reacquainted after a long period of time (I was better with technical). So we’re really delving into poorer note structure territory. This site is legitimately my notebook so pardon the extra work.
So we’ve constructed a number of fundamental models. We have various formulae and Python functions to analyze business. So now what? We want to be able to (subjectively) conclude the value of a stock out of a number of seemingly disparate maths.
From my primitive understanding we want to do a few things:
- Look historically at datapoints across balance sheets, income statements, cashflow, and outputs of our own formulaic calculations.
- Forecast values out multiple years.
- Add limiters and conditionals based on industry, mechanics of the business, and even empirical bias.
- Take those modified forecasts and reduce that back to a value today (Time value of money).
- Compare that to price per stock, stock cost, etc.
Now my assumption is I need to tackle a few new topics:
- I need to get some form of tabular data exported from this (excel).
- I need to find a way to access historic stock data (for free).
- I need to actually understand the materials!
- I’ll need some form of supervised machine learning model that can take in numeric data points. Maybe the historic data is so limited we just use a basic linear regression model first. Dunno. Hopefully one of my books will provide and I can go from there.
Valuation: Measuring and Managing the Value of Companies (Wiley Finance)
Amazon.com: Valuation: Measuring and Managing the Value of Companies (Wiley Finance): 9781119610885: McKinsey & Company Inc., Koller, Tim, Goedhart, Marc, Wessels, David: Books
For the actual modeling I think I’m going to have a deep reliance on the book “Valuation.” It boggles my brain a bit but it seems like I need to deeply understand chapters 10 - 28 with a lot of the heavy lifting coming from chapter 13, “The Mechanics of Forecasting.”
Understanding and integrating the concepts will be difficult, moreso automating the process. Why do I enjoy automation and coding? This would all be easier and well accepted to do purely in Excel.
- Programming is one of my best mediums to learn complex topics. Really learn them.
- You need to truly understand a business to valuate it. But automating some calculations saves a ton of time.
- I’m not doing this for anyone else, it’s not my job, and it certainly isn’t going to win me accolades. It’s fun, it’s a hobby. That grants me the sheer freedom to do whatever I want. Mwha-ha.
Valuation Research Notes
Enterprise Discounted Cash Flow
First we discount the free cash flow (FCF) at the weighted average cost of capital. And subtract debt.
- Value company operations by discounting free cash flow at the WACC.
- Cash flow generated by operations - reinvestment into business, discounted by WACC over periods.
- Value nonoperating assets like excess cash, marketable securities, nonconsolidated subsidiaries, and other non-op assets not in cashflow. Sum the value of operations and non operating assets to get the enterprise value.
- Value debt and nonequity claims: unfunded pension liabilities, employee options, preferred stock, etc.
- Subtract the value of debt from enterprise value to create the value of common equity. Then divide by current shares outstanding to estimate value per share.
- Reorg financial statements. Collect historic balance sheets. Calculate WACC and ROIC.
- Analyze historic performance. Focus on growth, ROIC excluding goodwill. ROIC.
- Project Rev Groeth, ROIC, and Free Cashflow. Build forecast model.
- Estimate continuing value.