Figure 5 - uploaded by Mathieu Bouville
Content may be subject to copyright.
Probability of real loss for a stock investment as a function of the investment length (in years). The gray straight line is a linear regression.
Source publication
Claims that the stock market is an investment for the long term are common, details on how long it takes less so, and empirical justifications are downright rare. In order to avoid losing purchasing power on one’s stock investment, it must last two decades (with regular investments, it takes around half a dozen more years). The uncertainty of retur...
Contexts in source publication
Context 1
... as mentioned in the previous paragraph, this number is based on only a few events and could a priori be pushed back by a long bear market (as was pointed out in the introduction). Figure 5 shows the probability of a real loss as a function of the length of the stock invest- ment. The risk of losing purchasing power did go down when investment duration increased: if one invested in stocks for two years the odd of losing money was one in four, but only a tenth of that over sixteen years. ...
Context 2
... the probability was historically exactly zero). This result is more robust than what was found with bear markets alone since it includes more numerous and more diverse episodes. The bump at 10-12 years is consistent with Figs. 2(a) and 3(a) showing several crashes needing a dozen years for recovery. The linear regression (gray straight line) in Fig. 5 is zero after 18.5 years. Since this is in the interval [18, 20.5] found above, we can say that if the long term means that one will not lose money then it takes between 18 and 20.5 years. Nonetheless, it must be remembered that such a duration is merely sufficient to avoid losing money, which is certainly not the goal of long-term ...
Context 3
... enough to show subtle cusps. This is perhaps also perhaps due to the model being 'too perfect': the uncertainty of 33-year or 66-year investments is unduly low in the model. Figure 17 shows the probability of losing money with the model. It goes down up to about twenty years then it stays low for a few years instead of going to zero, as it did in Fig. 5. The curve is smoother than with the S&P 500, since it is averaged over a hundred periods rather than just four. The off-diagonal terms of the correlation matrix are around a half on average, which is greater than those of Table 5 (0.3), as can be expected with a model which is more regular than reality. Overall, Figs. 15-17 show that ...
Context 4
... greater than those of Table 5 (0.3), as can be expected with a model which is more regular than reality. Overall, Figs. 15-17 show that a simple stochastic model with a succession of bull and bear markets can produce a great deal of what was observed with the S&P 500, including the flatter episodes in Figs. 13 and 10 and the extra time needed in Fig. 5 for the risk of losing money to reach zero. The model does not prove that these features are due to a succession of bull and bear markets, but it does show that such an explanation is ...
Similar publications
Recently, (Blanchet, Kang, and Murhy 2016) showed that several machine learning algorithms, such as square-root Lasso, Support Vector Machines, and regularized logistic regression, among many others, can be represented exactly as distributionally robust optimization (DRO) problems. The distributional uncertainty is defined as a neighborhood centere...
The “High North” is a place defined by constant change & uncertainty; this region has never before been integrated into regular shipping routes due to very harsh and unpredictable environmental conditions. Ice-breakers will be increasingly needed to smash through sea ice, open routes, or assist trapped vessels...
In order to complete nonlinear path following smoothly and accurately, this paper proposes a method utilizing the Serret-Frenet Line-of-Sight (LOS) guidance with adaptive compensation in the horizontal plane. All regular paths are feasible. Our method takes three steps to accomplish the path following. First, the guidance law calculates the desired...
