Escolar Documentos
Profissional Documentos
Cultura Documentos
models to
simulate the
future
A simple, multivariate model was
in the future under randomly generated states of the
constructed to correlate silver prices
economy.
with several economic indicators.
A statistically significant linear correlation exists
These included the rate of inflation,
between the silver price and the CPI. However, the price
the nominal level of the CPI and a
data is poorly represented by a simple linear correlation.
Other correlated factors must be included that explain 1970s and early 1980s. A real-world model would also
the obvious divergence of the data from the linear $/oz incorporate variables such as the elasticities of supply
vs. CPI trend line. The addition of other factors is re- and demand, inventory levels, consumption
ferred to as multivariate modeling. It provides a better expenditures and other variables that have known
overall fit with the real world data. It also enhances the correlations with the silver price. The multivariate
ability to forecast the impact of future states of the analysis of historical data from 1915 to 1995 resulted in
world on the price of silver. a functional model capable of predicting silver price with
Multivariate regression analysis indicates that the an adjusted RZ of 96 % for the period. The fit of the
price of silver at any point in time also depends on the model is shown on Fig. 4, which compares the actual
rate of change of the CPI or the level of inflation in the price data with the calculated data.
economy. The future cash flow from a silver producing The Markov-Monte Carlo inflation model and the
operation depends on the future price of silver. So fore- multivariate model can be combined to simulate likely
casts for capital budgeting must take potential inflation future states of the economy and the potential effect on
into account. the future price of silver. The top line of Fig. 5
Unfortunately, forecasting future inflation or defla- represents one potential scenario of inflation for the
tion is difficult. Inflation is caused by several economic next 25 years. It is derived from the Markov-Monte Carlo
mechanisms, or decisions, that impact the management inflation model seeded with 1995 inflation data. The
of growth of the money supply. From deficit spending bottom line of Fig. 5 presents the projected silver price
during times of war or to ease the effect of oil supply based on the above potential inflation projections.
shocks, inflation can be traced to monetary factors. If the actual future values of inflation and price are the
However, natural economic equilibrium constraints same as predicted, it will be accidental. However, the
function in the money markets, as in other markets. simulations represent unbiased projections of potential
This limits the duration and magnitude of sustainable states of the future. These are consistent with historical
inflationary episodes. experience and conditions of market equilibrium.
Figure 2 compares historical inflation data from Multiple simulations can be generated to provide cases
1950 to 1995 (points) with a correlation generated from for sensitivity analysis, to identify fatal scenarios and
a Markovian process model (solid line) in which inflation quantify the potential downside risk for capital bud-
in period, t, is linearly correlated with inflation in geting decision making. Worst- and best-case scenarios
period, t-1. The Markov correlation can be combined can be identified. Alternative planning for various situ-
with a Monte Carlo procedure to develop conditional, ations can be stimulated. Unbiased estimates of future
random estimates of the future path of inflation. cash flow also can be provided.
Figure 3 compares the actual inflation data and a All economic entities use financial forecasts to raise
Markov-Monte Carlo simulated inflation, seeded with capital, plan operations and make investments. Ulti-
the 1949 inflation rate, from 1950 to 1995. The data mately, the value of the company is determined by the
demonstrate that exact predictions of inflation are not value of the budget information and its correlation with
possible. However, multiple simulations of randomly the actual state of the economy in the future.
Simulation can provide glimpses of the most likely
future and quantify the probability of the occurrence of
individual states and the range of potential states.
4 SEPTEMBER 1997 MINING ENGINEERING
Management must interpret the results, plan for
contingencies and maximize the net present value for
the owners of the firm.