Nokia is one of those global companies that market watchers like to describe as a "bellwether stock".

Events that affect the Finnish mobile-phone maker often have implications for the whole technology sector; indeed, markets have sometimes taken their (short-term) direction from announcements made by Nokia.

The fact that the company has such a disproportionate impact on other stock prices is testament to the weirdness, at times, of markets. Nokia's share price jumped roughly 10 per cent in a couple of days thanks to last week's small upward revision to the firm's own sales and profits forecasts - another formidable challenge to those who argue that markets are essentially rational and forward-looking beasts.

One way or another, most of us own Nokia: it is one of those favourite stocks of individual and institutional investors alike. Its performance over the past few years provides an object lesson in market history: back at the peak of the bubble in early 2000, Nokia's share price was over five times today's level, which now stands at roughly the same level seen at the beginning of 1999.

That same description could, of course, be applied to countless other stocks; investors in Nokia, and other technology companies, will be hoping that earlier lows in share prices will not be revisited. In Nokia's case, that low was just under €9, compared to just over €11 at the time of writing.

One important lesson from those earlier lofty price peaks is that even sensible analysts can make whopping errors of judgment.

We know from the various investment scandals unearthed in the US that not all technology analysts believed what they wrote back then, but it is also important to remember that many of us were victims of irrational exuberance: opinions were honestly held but they were truly nonsensical. In researching this article, I came across many examples of the genre.

First, in January 2000 one analyst wrote: ". . . the p/e ratio is on the high side at 97.8 vs the industry average of 62.8 . . . price to sales looks good at 12.7." We can only wonder at the credulity of some people: currently, price to sales stands at just under 2 (although the analyst in question did warn, ultimately, that investors in Nokia should "proceed with caution").

Later in that same year, one of my favourite web sites, fool.com, reiterated its buy recommendation on the stock following a 25 per cent drop in the share price, arguing that sales and cash-flow growth was likely to exceed 30 per cent annually for the coming years; the stock subsequently fell by around 75 per cent over the next four years.

More recently, at the start of 2003, S&P awarded the company a "five-star investment rating", arguing the stock was 15 per cent undervalued to its discounted cash-flow based estimate of intrinsic value; that was when the stock was trading at around €16.

We all make mistakes, and before anyone starts digging into mine I put my hands up and admit to many errors of judgment. But we all need to remind ourselves of how recent it is that so many people got it so wrong.

Part of any investment case for Nokia today must be a belief that recent price lows represent a reaction to that earlier optimism and that the pendulum has swung too far the other way.

An orthodox analysis of Nokia's most recent income statement and balance sheet suggests the company is now in a strong financial position and is reasonably valued. There is very little debt on the balance sheet and a healthy pile of cash. Actually, the company generates bundles of cash; at around 10-11 times cash-flow, the current share price doesn't look too demanding.

Other simple metrics send similar stories: profit margins seem to be holding up better than some sceptics had feared; the dividend yield is healthy, particularly by technology sector standards. At around 14, the venerable p/e ratio also looks reasonable.

More complex discounted cashflow exercises paint a similar picture, although one that throws into sharp relief the critically important question of growth: what kind of sensible expectations can we form about Nokia's future earnings potential?At this point, we must confront one of Nokia's key problems, perhaps the main driver of its share price weakness: declining total sales revenues.

Earnings have held up better than sales because of cost controls. Correspondingly, its share of the global handset market has been declining, following a peak in 2002. Well-publicised problems with the design of its phones combined with other competitive pressures to produce a barrage of negative publicity.

Sceptical conventional wisdom argues that these trends will continue: Nokia's dominant market position can only be eroded and the market itself is becoming increasingly "commoditised" - which means only one thing, namely, downward pressure on margins, a problem for many in the technology sector.

Moreover, global saturation of the handset market, particularly in Europe and the US, means that sales growth will be increasingly hard to come by and all those hopes pinned on Asia (China in particular) will come to naught.

What is the ordinary investor to make of all of this? To buy Nokia shares now is essentially a bet that last week's positive trading statement is the start of a trend.

Nokia must deliver on its ambitious plans to expand the global handset market from 1.4 billion users to 2 billion by 2007 - with most of those extra customers coming from Russia, China, India, Brazil, Africa and parts of the Middle East.

The investor who is a non-specialist in mobile telephony can only guess at the achievability of these plans. Given their track record, I'm willing to bet that the experts' guesses are not wholly reliable either.

Nokia undoubtedly carries a "high risk" tag but that should not automatically exclude it from portfolios. Investors willing to take that risk would do well to look at a stock that has suffered from extremes of negative sentiment but nonetheless looks financially strong.

I started my research for this article as a sceptic: Nokia's financial strengths prompted some rethinking. Anyone willing to accept the company's own growth projections - always a difficult thing to do - would regard this stock as an aggressive buy.