Search

We propose a new methodology for estimating the demand and cost functions of dierentiated products models when demand and cost data are available. The method deals with the endogeneity of prices to demand shocks and the endogeneity of outputs to cost shocks, but does not require instruments for identication. We establish non-parametric identication, consistency and asymptotic normality of our estimator. Using Monte-Carlo experiments, we show our method works well in contexts where instruments are correlated with demand and cost shocks, and where commonly-used instrumental variables estimators are biased and numerically unstable.

With consideration to the Curie-Weiss law, we investigate magnetization reversal beyond the Curie temperature, from theoretical to experimental. This paper outlines a theoretical model in accordance with the Curie-Weiss law and also presents a new criterion for the shape stability of recorded domains using magneto-optical recording or hybrid recording. Fairly good agreement was obtained between the theoretical and experimental investigations.

In this paper, we estimate the effect of workfare policy on crime by exploiting two exogenous welfare policy changes in Denmark. Our results show a strong decline in the crime rate among treated unemployment uninsured men relative to untreated uninsured and unemployment insured men, and part of this decline can be identified as a direct effect of workfare participation. Moreover, we find that criminal activity was also reduced during weekends, when the workfare programs were closed, allowing us to distinguishing the pure program effect from the incapacitation effect. These results imply a strong and potentially lasting crime reducing effect of workfare policy.

本稿は、もと学会（日本基督教学会近畿支部会一九九七年三月）における発題講演の原稿に、要望にしたがい、倉卒の間に手を入れ、まとめられたもの

The behavior of the air flow inside a hard disk drive is evaluated by a three-dimensional unsteady-flow analysis technique called large eddy simulation (LES). The simulation shows that the fluctuation of the aerodynamic force is strongly correlated with the amplitude of the disk vibration, which causes positioning error of the read/write head. It also shows that the power loss caused by the friction of the air on the disks, which accounts for most of the power consumption of hard disk drives at a high-speed revolution, can be qualitatively predicted.

The combination of Magnetic Amplifying Magneto-Optical System (MAMMOS) media, a blue laser and a high NA objective lens has been investigated using simulations and experiments. This paper describes the experimental and simulated results of domain copying phenomena for the MAMMOS media using a 413nm blue laser and an objective lens with a numerical aperture (NA) of 0.9. Without an external magnetic field, the readout laser was irradiated onto the MAMMOS media. Since the phenomena were related to Rear Aperture Detection phenomena, the signal amplitude simulation assumed temperature dependence of the Kerr rotation, the aperture thermal properties and the Gaussian laser beam profile. This paper also presents the analysis of expansive and shrinking forces per unit wall area concerning the MAMMOS readout using the blue laser and the high NA objective lens.

Very small domains are recorded on the MAMMOS disk and they ran be detected using RF readout field. The MO signal amplitude of 20 nm domains can be enhanced to the saturated signal level. For these small domains, the thermal stability problem is investigated. Even if high laser power for readout is irradiated to the small domains, the reproduced signal was maintained. The result indicates that the recording layer of TbFeCo possesses an advantage for paramagnetic limit problem over 100 Gbits/in/sup 2/.

For almost all optical disks, signal decline is mainly due to inter symbol interference (ISI) caused by laser beam spot size or magnetic aperture size. Such 151 does not appear for Zero Field Magnetic Amplifying Magneto-optical System (Zero Field MAMMOS). The limit of the frequency response comes from the frequency range of the MO readout channel. MAMMOS has a resolution for 1/10 of a laser beam corresponding to 0.1μm. Decrease in the carrier level and error of less than 0.1μm is caused by expansion failure. If expansion failure were circumvented, Zero Field MAMMOS would have a possibility of readout with 0.04μm resolution using a laser with wavelength, 650nm and an objective lens of NA, 0.6.

We examine the ability of male immigrants to transfer their occupational human capi- tal using information from the O*NET and a unique dataset that includes both the last source country occupation and the first four years of occupations in Canada. We first augment a model of occupational choice and skill accumulation to derive predictions about the cross-border transferability of occupational human capital. We then test the empirical implications using the skill requirements of pre- and post-immigration occupations. We find that male immigrants to Canada were employed in source coun- try occupations that required high levels of cognitive skills, but relied less intently on manual skills. Following immigration, they find initial employment in occupations that require the opposite. Regression analysis uncovers large returns to the analyti- cal skill requirements of Canadian occupations, but no returns to source country skill requirements. Finally, our empirical findings suggests that occupational skill gaps are detrimental to immigrants' earnings.

This paper proposes a new test of the Protection for Sale (PFS) model by Grossman and Helpman (1994). Unlike existing methods in the literature, our approach does not require any data on political organization. We use quantile and quantile IV regressions to do so using the data from Gawande and Bandyopadhyay (2000). Surprisingly, the results do not provide any evidence favoring the PFS model. We also explain why previous work may have inadvertently found support for it., グローバルCOEプログラム = Global COE Program

In this paper, we focus on a novel and potentially important aspect of the workfare policy in the Danish labor market, namely its effect on crime. We do this by exploiting two policy changes. First, we examine the effect of a series of national welfare reforms introduced during the 1990s. Those reforms strengthened the work requirement for the young welfare recipients and were introduced gradually, starting with younger welfare participants first. We exploit the differential introduction of workfare reform across different age groups as the exogenous variation. Second, we use a unique policy experiment that began in 1987 by an innovative mayor of the Danish city of Farum, where he imposed a 100 % work or training requirement for all welfare recipients immediately from the date of enrollment. By comparing the changes in crime rates among the welfare recipients in Farum before and after 1987 with that of the rest of Denmark, we identify the effect of workfare on the crime rate.

This paper provides a step-by-step guide to estimating discrete choice dynamic programming (DDP) models using the Bayesian Dynamic Programming algorithm developed in Imai, Jain and Ching (2008) (IJC). The IJC method combines the DDP solution algorithm with the Bayesian Markov Chain Monte Carlo algorithm into a single algorithm, which solves the DDP model and estimates its structural parameters simultaneously. The main computational advantage of this estimation algorithm is the efficient use of information obtained from the past iterations. In the conventional Nested Fixed Point algorithm, most of the information obtained in the past iterations remains unused in the current iteration. In contrast, the Bayesian Dynamic Programming algorithm extensively uses the computational results obtained from the past iterations to help solving the DDP model at the current iterated parameter values. Consequently, it significantly alleviates the computational burden of estimating a DDP model. We carefully discuss how to implement the algorithm in practice, and use a simple dynamic store choice model to illustrate how to apply this algorithm to obtain parameter estimates.

This paper provides a step-by-step guide to estimating discrete choice dynamic programming (DDP) models using the Bayesian Dynamic Programming algorithm developed in Imai, Jain and Ching (2008) (IJC). The IJC method combines the DDP solution algorithm with the Bayesian Markov Chain Monte Carlo algorithm into a single algorithm, which solves the DDP model and estimates its structural parameters simultaneously. The main computational advantage of this estimation algorithm is the efficient use of information obtained from the past iterations. In the conventional Nested Fixed Point algorithm, most of the information obtained in the past iterations remains unused in the current iteration. In contrast, the Bayesian Dynamic Programming algorithm extensively uses the computational results obtained from the past iterations to help solving the DDP model at the current iterated parameter values. Consequently, it significantly alleviates the computational burden of estimating a DDP model. We carefully discuss how to implement the algorithm in practice, and use a simple dynamic store choice model to illustrate how to apply this algorithm to obtain parameter estimates.

There is little work on the inner workings of journals. What factors seem to affect the ability to publish in a journal? Could simple rules (which are already used by some journals) like the immediate rejection of a significant minority of papers, help to streamline the process? At what cost? How well do journals seem to do in choosing papers? What can we say about the extent of type 1 and type 2 errors? Do editors seem to have uniform standards or are some harsher than others? We use data on submissions to the Journal of International Economics to help answer these questions.

This paper critically and selectively surveys the literature on protec- tion for sale and discusses directions for future research in this area. It suggests that the standard approach need to be augmented to provide more compelling tests of this model.

This paper proposes a new test of the Protection for Sale (PFS) model by Grossman and Helpman (1994). Unlike existing methods in the literature, our approach does not require any data on political organizations. We formally show that the PFS model provides the following prediction: In the quantile regression of the protection measure on the inverse import penetration ratio divided by the import demand elasticity, its coefficient should be positive at the quantile close to one. We examine this prediction using the data from Gawande and Bandyopadhyay (2000). The results do not provide any evidence favoring the PFS model.

We propose a new methodology for structural estimation of dynamic discrete choice models. We combine the Dynamic Programming (DP) solution algorithm with the Bayesian Markov Chain Monte Carlo algorithm into a single algorithm that solves the DP problem and estimates the parameters simultaneously. As a result, the computational burden of estimating a dynamic model becomes comparable to that of a static model. Another feature of our algorithm is that even though per solution-estimation iteration, the number of grid points on the state variable is small, the number of effective grid points increases with the number of estimation iterations. This is how we help ease the "Curse of Dimensionality". We simulate and estimate several versions of a simple model of entry and exit to illustrate our methodology. We also prove that under standard conditions, the parameters converge in probability to the true posterior distribution, regardless of the starting values.

This paper asks whether the results obtained from using the standard approach to testing the in‡uential Grossman and Helpman “protection for sale”model of political economy might arise from a simpler setting. A model of imports and quotas with protection occuring in response to import surges, but only for organized industries, is simulated and shown to provide parameter estimates consistent with the protection for sale framework. This suggests that the standard approach may be less of a test than previously thought.

~, ~~ the future of high density MO recording. The problem should be investigated not only for HDD, but also for MO. There are few reports concerning this thermal stability problem for MO. The reason is due to the difficulty in detecting MO signals for the very tiny domains involved, less than 0.1 pm. As one solution for this small domain detection, we proposed a magnetic domain expansion detection named MAMMOS (Magnetic AMplifying MO System). By using this technique, very tiny MO signals reproduced from small domains less than 0.1 vm in length can be enhanced to the saturated signal level”. ’I. In this paper, the thermal stability of these very tiny domains is discussed. Experimental Drocedures The readout layer (GdFeCo:20nm) and recording layer (TbFeCo50nm) were prepared by a magnetron sputtering process on a Poly-Carbonate substrate. The typical TbFeCo for MO recording layer is an amorphous rare earth transition metal alloy. The coercive force at room temperature is quite large (over 20kOe). From this characteristic, it seems that very small domains could be recorded very stable. However, the experimental data were not reported. A non-magnetic intermediate layer separated each magnetic layer. In order to investigate the thermal stability for very small domains, a read-write tester was used with 680nm wavelength, with an objective lens of numerical apeaure, 0.55. The linear disk velocity was 0.8dsec. Exoerimental results and discussions Generally, high readout laser power irradiation causes reformation of the recorded domains because of decreased thermal stability of the domains. This leads to data error. We investigated the change in readout signal before and after irradiation of high laser power to the recorded areas in a MAMMOS disk. In one example, repetition patterns of 0.031.” (30nm) domains and 0.97pm spaces were recorded and the readout waveforms, observed. After this, the high lascr power was irradiated several times onto the recorded areas and the readout waveforms were observed again. In this way, we measured the irradiation number of the high laser power until an error MAMMOS signal was observed. The relation between the laser power and maximum temperature of the recording layer in the laser spot was estimated. According to this, the readout laser powers were transformed to the maximum temperature of the laser spot. The resulting thermal stability is shown in Fig.]. This is a typical graph of Anhenius plot to illustrate the thermal stability. The plots show a linear relation between the inverse temperature and the number of high power laser irradiation corresponding to when error a peared. In the case of a MAMMOS readout temperature region around 1.97e-3 K- , the laser P irradiation number indicates over le10. This is sufficient for actual data reproduction. Also, in the region of 373K (lOOC), the thermal stability of 0.03pm domains was investigated. The laser irradiation number was over le50. This number corresponds to at least over 100 years of archival data storage. In the figure, there are two symbols. The closed squares correspond to the result of the 0.03pm domains and the triangles are those of the 0.10 pm domains. However, there is no apparent difference between them. Conclusion Thermal stability of 0.03pm recorded domain on the MAMMOS disk was investigated. The results indicated that the readout cycle is over le10 times and the archival life at around lO0C is beyond 100 years

Summary form only given. Hybrid recording has already been realized in laser beam assisted perpendicular magnetic recording in audio recordable MiniDiscs. In the year 2000 alone, more than 20 million drives and 220 million disks were shipped. Hybrid recording is expected to be a very strong candidate in the near future in the very high density region, exceeding terabits/in/sup 2/. Here, we report on hybrid recording with the capability to overcome two of the main serious problems; the paramagnetic limit in high density region and the writing speed limit caused by ferromagnetic resonance around 2 to 5 GHz.

Hybrid recording is a novel scheme combining optical recording and magnetic recording (H. Saga et al, J. Magn. Soc. Jpn, vol. 23, suppl. S1, p. 225, 1999). In this paper, basic experiments satisfying the principles of both very high density and ultra high speed writing are shown. We discuss the future capability of the storage device toward Tbit/in/sup 2/ and 100 Gbps directly coupled to an optical fiber communication network.

To obtain an areal density of 100 Gbit/in/sup 2/, magneto-optical (MO) technologies are attractive from the perspectives of magnetic field modulation recording, magnetic domain expansion readout, and high read/write cyclability. In this paper, we discuss the micro magnetic mechanism and compare the domain expansion MO schemes.

Magnetic Amplifying Magneto-Optical System (MAMMOS) shows human brain like memory behavior. Magnetic field and laser power have threshold to recover the stored memory like the human response of remembering. MAMMOS also has a feature to amplify very small recorded signals like our recovery of memory, e.g. fifty years ago episode.By adding the meaningful information on the magnetic field pattern, we can get some correlation between our memory and external stimulation. Such scheme is named as “the Active readout MAMMOS” which is analogues to the human process of remembering the memory.If the applied field pattern and timing phase just coincide with stored information, there occurs the coherent amplification of MAMMOS signal. We can utilize such phenomena as the trigger of “Memory Association”.