頁籤選單縮合
| 題 名 | 以個別公司財務報表資訊測試生產平穩化理論 |
|---|---|
| 作 者 | 李文智; 李淑華; 蔡彥卿; | 書刊名 | 管理與系統 |
| 卷 期 | 2:1 1995.01[民84.01] |
| 頁 次 | 頁85-100 |
| 分類號 | 494.542 |
| 關鍵詞 | 生產平穩化; 存貨不足; 財報資訊; 在製品; Production smoothing; Stock out; Disagregated data; Work-in-process; |
| 語 文 | 中文(Chinese) |
| 中文摘要 | 本文以動態規畫模型(dynamic programming model)解釋生產、銷貨及存 貨變動等三個變數的變異及共變異數的結構。吾人首先以損益表的恆等式證明 存貨變動的衡量必須包含在製品，其次，再說明此類研究中銷貨之衡量必須以 銷貨成本為之。在這些修正之下，證明(1)生產的變異數大於銷貨的變異數，及 (2)銷貨與存貨變動的共變異數應為負數。此一模型與其他類似研究之差異如下: 1.過去之文獻均以總體(產業)資料驗證這一類生產平穩化模型(production smoothingmodel)。然而，這類模型屬於個體廠商之模型，理應以個體資料加以 驗證。本文以個別公司的損益表為出發點，使吾人得以利用個別廠商的損益表 及資產負債表做實證研究。2.過去的經濟文獻大多忽略在製品，本文證明此一 疏忽造成偏差的估計。3.以個別廠商的資料使吾人得以改善對銷貨衡量的準確 性。收集1980至1990年間，我國上市公司中的水泥、食品、玻璃、造紙、汽車 及建築等六個產業(29家公司)的資料做實證研究，結果強烈支持生產平穩化理 論(the productionsmoothing theory)。六個產業中每一個產業的平均變異數比率 (average varianceratio)均小於1。25家公司的銷貨與存貨變動的共變異數為負數。 最後，吾人將個別廠商之資料加總，證實Lai[16]所言「加總後之資料會使結果 產生偏誤」之說法應是正確的。 |
| 英文摘要 | In this paper, we apply a dynamic programming model to explain the variance-covariance structure of production, sales, and changes in inventory level for manufacturing firms. We first demonstrate, through the identity on income statements that work-in-process should not be omitted in the calculation of inventory and that cost of good sold, instead of sales, should be used to measure shipment. We then show that variance of units produced (manufacturing cost)should be smaller than the variance of units sold (cost of goods sold). Also, the covariance between cost of goods sold and change in inventory, both in physicalunits and in dollar term, should be negative. There are three distinct features of this model:1. It enables us to utilize financial statements as a competing data resource to test for the production smoothing model. This is of importance, because the production smoothing model is a microeconomic model and consequently, the use of macroeconomic data (at industry level) in the tests causes the well known aggregation bias.2. Work-in-process is ignored in most economic literature. But, it is proved in our model that the omission produces another source of bias.3. Similarly, cost of goods sold replaces sales in the traditional models of production smoothing that adjusts for the fact that "one dollar of sales represents less physical units than one dollar of inventory". We examine twenty nine firms in cement, food, glass, pulp and paper, automobile and construction industries in Taiwan for the years from 1980 to 1990 and our results strongly support the production smoothing theory. The average variance ratios is smaller than unity for allsix industries in our sample. The covariances are negative for twenty five out of the twenty nine firms. These results strongly support the production smoothing model. In addition, we form data at industry level by pooling data across firms within industries and find that aggregation may (as Lai[16] hypothesized) lead to biased inference. Cement, food, paper, and construction industries are examined in this process. Within the four industries, the variance ratios of paper and construction industries are larger than unity and those of cement and food are still less than unity. |
本系統中英文摘要資訊取自各篇刊載內容。