...

In Graham and Krugman(1995), foreign direct investment is formally defined as ``ownership of assets in one country by residents of another for purposes of controlling the use of those assets.'' See Graham and Krugman(1995) for more details and the concerning issue.

...

The coefficients of correlation is -0.7384.

...

Froot(1993) provides a wide range of empirical and policy studies on FDIs.

...

Due to the shortages of the data of Japan's FDI flow to the United States, the Japan's FDI to North America is examined in this paper. The empirical study of the Japan's FDI in the U.S. will be conducted in the near future.

...

See Yoshikawa(1992) and Yoshikawa(1995).

...

What is meant by ``representative'' comes from the idea that macroeconomic phenomena can be explained by behavior of the ``representative'' microeconomic agent. Of course, what we see as FDI is a result of numerous multinational firms' behavior. But it is out of question to examine the behavior of each multinationals. For analytical handlefulness, all multinationals are assumed to act as if they were `one' multinational firm.

...

See Lucas(1967), Gould(1968) and Treadway(1969) for more details about adjustment cost functions of investment. Since all adjustment costs are assumed to be borne by home firms as described in (A8), both adjustment functions ¦µ(ˇ¦) and ¦¨(ˇ¦) are introduced in the home firm's cost functions, as in (3-2-2-c)

...

In the following empirical study, variables in (5-1-1) have the following units. Y(unit:billion yen), K(billion yen), L(ten thousand workers), h(hours/a workerˇ¦a year). It is, of course, totally up tp empirical researchers what unit to take. One must keep in mind that changes in unit of these variables will affect the values of technological level(A) in (5-1-1) to a great extent.

...

It is very important in empirical studies using production functions to examine what data should be used as factor inputs. Kuroda(1984) and Kuroda(1989) show how to make capital- and labor input data in estimating production functions. Kuroda also stresses the importance of building and dealing data carefully.

...

Some would ask what the nominal wage rate in North America is. Of course, there is not such things. But conceptually such thing as a nominal wage rate in North America can be introduced in theoretical model. Some problems arising from this will be discussed later.

...

This assumption is somewhat arbitrary, which makes this paper's quality a bit lower. A more accurate estimating of the capital's lifetime is necessary and is now in progress.

...

Using the data calculated by (7-1-1) will make this study more credible. But if we use this method, this study is no longer a macroeconomic research. It is likely to be called an industrial-organization research.