Should you be looking to the past to predict the future when it comes to sales trends? Certainly those people affected by various economic slumps in the past would suggest that relying on historical data to create your financial model of the future is a mistake, but other economists and market experts believe historical data gives us very strong indications of what to expect in the future. When you use a financial model it requires assumptions about the underlying assets. The two main assumptions are the assets’ expected price and the impact of market volatility. To arrive at the assumptions that then inform market price, you can be guided by either data collected from previous years or personal opinions and viewpoints. It is a case of objective-based judgement versus subjective-based judgment. Personal opinions are guided by instinct and experience and, as such, must be seen as unreliable and hardly the basis for credible business decisions. When it comes to personal opinion, there is a temptation or a tendency to use assumptions that make your product most attractive. If times are bad, the market will question such optimism but if the economy is experiencing a high, then your optimistic outlook might get overlooked as the trend is an upwards and positive one – in other words, your company may still be making money, just not as much as it could. Historical data, on the other hand, may have its issues but it is the only objective way to measure risk. One question you will be asking yourself is how far back should you mine data? Certainly this is something to be wary of. The bewildering speed of technological change, globalisation, policy shifts and the instability of world markets and their knock-on effect can make using data more than a couple decades old inappropriate. The data will just not be meaningful enough in the face of so much change. Take the toxic example of house prices. The house price crash in 2007 and 2008 offers some insight into this. Many people thought the UK’s housing market was in a bubble, yet the data used by many financial engineers did not include a significant, nationwide fall in house prices. Such data did not exist, unless you went back to the 1930s. However, while it is difficult to know the extent to which you will use historical data, you still need to give your assets a price in order to trade. The price of an asset is a function of its expected return and risk. So anytime you trade an asset, you're taking a position on its future value. Markets are more liquid when there is a consensus on prices and freeze when no one can agree. When asset markets freeze credit becomes scarce and economic growth stalls. Risk modelling, based on historical data, lubricates financial markets by giving traders an objective guide they can all use. So yes, using historical data is not the perfect solution but it does remain the most unbiased way to measure risk and make assumptions about the future. For today’s business analysts, the skill will be in gaining enough information from historical data to allow them to judge what data is useful and what should be consigned to history.