I have been following Nvidia for several years now and have always asked myself what it takes for Morningstar to revisit or review a valuation model, or at least challenge the analysts on the validity of the assumptions.
For some of us value investors, it is clear that the market is not always efficient, and there will be times, sometimes short, where the value of a stock is in disagreement with its price. I guess Morningstar is all about trying to understand those opportunities the market presents where stock are all of a sudden undervalued and present a good entering opportunity.
Sometimes the fair value lies above, sometimes below the last close....but, I find it hard to make sense out of the history behind the Nvidia valuation; How can the valuation have been so off since 2015? Its been 5 years lagging the market, while the stock evolved from 21USD to their last close of 420!!!!! The fair value always almost 50% below, but still being adjusted (late) in a very shy fashion.
It feels like everybody saw the potential and development of this company, but the only ones that never were able to make money on it were the ones following Mr Davuluri´s analysis.
I understand , the market can be wrong, now and then, and sometimes for a longer period of time. But, when is the time to look back inside an reflect? are all the participants in the market wrong? or is there something flawed in my model? Especially when I call my uncertainty "very high" I would expect to go back and reflect on how good, this very highly uncertain model has performed, and what that performance has meant for the customers of my analysis.
Just speaking theoretically, a model is only as good as its profile vs. reality. My naive interpretation of events would lead me to the conclusion that the model is badly flawed in some way. It seems as if there must be some additional factor(s) which the model does not reflect. The choices would be to: abandon the model, identify and incorporate the missing factors, OR ascertain under what set of circumstances the model will and will not apply. Continuing to use the model as is does not seem to be a good choice to me. I realize that's not exactly what you were going for, but that would be my knee-jerk reaction.
To your actual question, I'd be trying to revise my system NOW...
I believe a complete unknown future is even more unpredictable in the short term. For me short term is under 5-10 years or a couple of cycles. NVDA is a top niche momentum stock not a long term game changer at this point. More herd mentality may be involved which is not easily subject to numerical analysis akin to traditional value metrics. Apples to oranges.
I don’t believe Mr. Market is ever wrong. I do believe the herd is being led to believe tech, any tech, is going to lead an economic recovery by the financial gypsies. This may just be another example of market timing run amok. The model may eventually be correct when the herd is told to move on to something else.
High PE momentum stocks MSFT, AAPL in the 80’s and AMZN in the OO’s, were all game changers and turned out to have staying power. We just jumped on when the momentum took a breather.
*Take a look at SHW. Good company bad valuation. I can’t figure that one out other then DIY programming popularity.