Background: I'm a very small-time (eg. $100s), inexperienced trader. Currently, I'm buying and holding stocks that bear dividends.

When I evaluate stocks, I look at certain criteria; these have a specific purpose. For example:

  • Market cap must be $10B or more (price manipulation less likely)
  • Earnings per share exceeds dividends per share (sustainable dividends)
  • And more.

As I add more of these criteria, previous holdings of mine may not meet some of the new criteria.

My question is, how do I evaluate if I should keep holding and buying, or just keep holding, or liquidate my holdings?

Because my profile is conservative and long-term holdings, I probably don't want too many holdings that don't meet my requirements. But I'm not sure how to determine when I should stop buying, or liquidate. (I would keep buying these holdings otherwise to ensure that no single holding dominates my portfoio.)

Edit: My goals can be prioritized as (in order):

  • Build income generation through investment (eg. dividends)
  • Invest in something "stable" that will go up in value (even if the horizon is 10+ years)
  • 1
    The answer to your questions depends entirely on your investment strategy: what is your goal, and what are acceptable risks to achieving that goal? Buying stocks that issue dividends is a classic conservative investment strategy, but depending on your age and the relative value of your portfolio, buying non-dividend stocks may be only marginally less conservative with significantly greater reward.
    – Nick2253
    Mar 31, 2015 at 17:13
  • 1
    @Nick2253 I've updated my question to answer that. My goals are income generation, and then long-term growth.
    – ashes999
    Mar 31, 2015 at 19:37
  • You don't really address your risk profile. Depending on acceptable risk, dividend stocks may not be the best way to generate income.
    – Nick2253
    Mar 31, 2015 at 20:14
  • @Nick2253 maybe you can clarify: what do you mean by "risk profile," exactly?
    – ashes999
    Mar 31, 2015 at 20:36
  • 1
    @ashes999, position sizing will teach you how much of each stock you should buy in the first place, it has nothing to do buying on margin or with cash. It sounds like your strategy is not much of a strategy, because you are all over the place. A strategy should be telling you what to buy, when to buy, how much to buy (which is where position sizing comes in), when to sell, and it should have risk management incorporated into the strategy (what to do if things don't go how you expected them to go).
    – user9822
    Apr 10, 2015 at 23:20

3 Answers 3


Don't sell. Ever. Well almost.

A number of studies have shown that buying equal amounts of shares randomly will beat the market long term, and certainly won't do badly. Starting from this premise then perhaps you can add a tiny bit extra with your skill... maybe, but who knows, you might suck. Point is when buying you have the wind behind you - a monkey would make money.

Selling is a different matter. You have the cost of trading out and back in to something else, only to have changed from one monkey portfolio to the other. If you have skill that covers this cost then yes you should do this - but how confident are you?

A few studies have been done on anonymised retail broker accounts and they show the same story. Retail investors on average lose money on their switches.

Even if you believe you have a real edge on the market, you're strategy still should not just say sell when it drops out of your criteria. Your criteria are positive indicators. Lack of positive is not a negative indicator. Sell when you would happily go short the stock. That is you are really confident it is going down. Otherwise leave it.

  • 1
    Minor quibble: sell if necessary to maintain you target ratios between different kinds of investments ... but it may be better, when possible, to rebalance by changing what you're now buying.
    – keshlam
    Jun 1, 2015 at 14:41
  • @keshlam that is a good point - to the extend that you have portfolio level targets, then changing your buying rules is a much better way to change your portfolio that to actually switch out stocks.
    – Corvus
    Jun 1, 2015 at 14:43

If your criteria has changed but some of your existing holdings don't meet your new criteria you should eventually liquidate them, because they are not part of your new strategy.

However, you don't want to just liquidate them right now if they are currently performing quite well (share price currently uptrending).

One way you could handle this is to place a trailing stop loss on the stocks that don't meet your current criteria and let the market take you out when the stocks have stopped up trending.

  • This doesn't really answer my question. It says "yes you should eventually liquidate them" (which I know) and "one way is ..." which is not what I asked; I asked about how to tell when to liquidate.
    – ashes999
    Apr 10, 2015 at 18:28
  • 1
    @ashes999, if you have a certain criteria of what stocks to keep holding and that criteria changes, then if you keep holding stocks you bought previously under a different criteria, then you might as well be just holding anything in your porfolio. I have provided a emotion free way to eventually get out of these stocks and possibly make a profit from them as well. In other words the market will decide for you when is the right time to liquidate them, leaving your emotions out of it.
    – user9822
    Apr 10, 2015 at 23:12

Unfortunately I believe there is not a good answer to this because it's not a well posed problem. It sounds like you are looking for a theoretically sound criteria to decide whether to sell or hold. Such a criteria would take the form of calculating the cost of continuing to hold a stock and comparing it to the transactions cost of replacing it in your portfolio.

However, your criteria for stock selection doesn't take this form. You appear to have some ad hoc rules defining whether you want the stock in your portfolio that provide no way to calculate a cost of having something in your portfolio you don't want or failing to have something you do want.

Criteria for optimally rebalancing a portfolio can't really be more quantitative than the rules that define the portfolio.

  • Can you give an example of how I can use quantitative rules to define my portfolio better, and how I can use those to calculate the cost of keeping them in my portfolio?
    – ashes999
    May 1, 2015 at 14:25
  • Defining a portfolio is a very complex problem with lots of approaches so I'm not sure how good my example would be. Let's say you have a model giving you expected returns and covariance for your portfolio. Your target portfolio could maximize some criteria, like the sharpe ratio. You could compare your current portfolio under that criteria against your target minus the transactions cost. If the target minus transactions cost is better, it may make sense to make the trades.
    – farnsy
    May 1, 2015 at 14:56
  • That's all far beyond my current understanding. Maybe you can cite a couple of readings/links and I can pick it up from there.
    – ashes999
    May 1, 2015 at 18:29
  • To have a quantitative portfolio you would need to throw out everything you currently do and start fresh. I'm not particularly recommending that. Since your investment strategy is based essentially on what seems good heuristically, your rebalance strategy should be the same. Sell when it doesn't seem like a good idea to hold it any more. I know it doesn't help much, but that's why I said there is not a good answer to this question. Personally the tax implications are the main criteria I look at when I am deciding whether to sell.
    – farnsy
    May 1, 2015 at 23:36
  • I didn't say I'm necessarily going to restart quantitatively. I just want you to point me to some readings that teach and explain the concepts you're talking about.
    – ashes999
    May 2, 2015 at 12:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .